2020-01-02 16:52:30 +00:00
|
|
|
/*
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
* libADLMIDI is a free Software MIDI synthesizer library with OPL3 emulation
|
2020-01-02 16:52:30 +00:00
|
|
|
*
|
|
|
|
* Original ADLMIDI code: Copyright (c) 2010-2014 Joel Yliluoma <bisqwit@iki.fi>
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
* ADLMIDI Library API: Copyright (c) 2015-2020 Vitaly Novichkov <admin@wohlnet.ru>
|
2020-01-02 16:52:30 +00:00
|
|
|
*
|
|
|
|
* Library is based on the ADLMIDI, a MIDI player for Linux and Windows with OPL3 emulation:
|
|
|
|
* http://iki.fi/bisqwit/source/adlmidi.html
|
|
|
|
*
|
|
|
|
* This program is free software: you can redistribute it and/or modify
|
|
|
|
* it under the terms of the GNU General Public License as published by
|
|
|
|
* the Free Software Foundation, either version 3 of the License, or
|
|
|
|
* any later version.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License
|
|
|
|
* along with this program. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
*/
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
#include "adlmidi_opl3.hpp"
|
2020-01-02 16:52:30 +00:00
|
|
|
#include "adlmidi_private.hpp"
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <cassert>
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
#ifndef DISABLE_EMBEDDED_BANKS
|
|
|
|
#include "wopl/wopl_file.h"
|
|
|
|
#endif
|
|
|
|
|
2020-01-02 16:52:30 +00:00
|
|
|
#ifdef ADLMIDI_HW_OPL
|
|
|
|
static const unsigned OPLBase = 0x388;
|
|
|
|
#else
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
# if defined(ADLMIDI_DISABLE_NUKED_EMULATOR) && \
|
|
|
|
defined(ADLMIDI_DISABLE_DOSBOX_EMULATOR) && \
|
|
|
|
defined(ADLMIDI_DISABLE_OPAL_EMULATOR) && \
|
|
|
|
defined(ADLMIDI_DISABLE_JAVA_EMULATOR)
|
2020-01-02 16:52:30 +00:00
|
|
|
# error "No emulators enabled. You must enable at least one emulator to use this library!"
|
|
|
|
# endif
|
|
|
|
|
|
|
|
// Nuked OPL3 emulator, Most accurate, but requires the powerful CPU
|
|
|
|
# ifndef ADLMIDI_DISABLE_NUKED_EMULATOR
|
|
|
|
# include "chips/nuked_opl3.h"
|
|
|
|
# include "chips/nuked_opl3_v174.h"
|
|
|
|
# endif
|
|
|
|
|
|
|
|
// DosBox 0.74 OPL3 emulator, Well-accurate and fast
|
|
|
|
# ifndef ADLMIDI_DISABLE_DOSBOX_EMULATOR
|
|
|
|
# include "chips/dosbox_opl3.h"
|
|
|
|
# endif
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
|
|
|
|
// Opal emulator
|
|
|
|
# ifndef ADLMIDI_DISABLE_OPAL_EMULATOR
|
|
|
|
# include "chips/opal_opl3.h"
|
|
|
|
# endif
|
|
|
|
|
|
|
|
// Java emulator
|
|
|
|
# ifndef ADLMIDI_DISABLE_JAVA_EMULATOR
|
|
|
|
# include "chips/java_opl3.h"
|
|
|
|
# endif
|
2020-01-02 16:52:30 +00:00
|
|
|
#endif
|
|
|
|
|
|
|
|
static const unsigned adl_emulatorSupport = 0
|
|
|
|
#ifndef ADLMIDI_HW_OPL
|
|
|
|
# ifndef ADLMIDI_DISABLE_NUKED_EMULATOR
|
|
|
|
| (1u << ADLMIDI_EMU_NUKED) | (1u << ADLMIDI_EMU_NUKED_174)
|
|
|
|
# endif
|
|
|
|
|
|
|
|
# ifndef ADLMIDI_DISABLE_DOSBOX_EMULATOR
|
|
|
|
| (1u << ADLMIDI_EMU_DOSBOX)
|
|
|
|
# endif
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
|
|
|
|
# ifndef ADLMIDI_DISABLE_OPAL_EMULATOR
|
|
|
|
| (1u << ADLMIDI_EMU_OPAL)
|
|
|
|
# endif
|
|
|
|
|
|
|
|
# ifndef ADLMIDI_DISABLE_JAVA_EMULATOR
|
|
|
|
| (1u << ADLMIDI_EMU_JAVA)
|
|
|
|
# endif
|
2020-01-02 16:52:30 +00:00
|
|
|
#endif
|
|
|
|
;
|
|
|
|
|
|
|
|
//! Check emulator availability
|
|
|
|
bool adl_isEmulatorAvailable(int emulator)
|
|
|
|
{
|
|
|
|
return (adl_emulatorSupport & (1u << (unsigned)emulator)) != 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
//! Find highest emulator
|
|
|
|
int adl_getHighestEmulator()
|
|
|
|
{
|
|
|
|
int emu = -1;
|
|
|
|
for(unsigned m = adl_emulatorSupport; m > 0; m >>= 1)
|
|
|
|
++emu;
|
|
|
|
return emu;
|
|
|
|
}
|
|
|
|
|
|
|
|
//! Find lowest emulator
|
|
|
|
int adl_getLowestEmulator()
|
|
|
|
{
|
|
|
|
int emu = -1;
|
|
|
|
unsigned m = adl_emulatorSupport;
|
|
|
|
if(m > 0)
|
|
|
|
{
|
|
|
|
for(emu = 0; (m & 1) == 0; m >>= 1)
|
|
|
|
++emu;
|
|
|
|
}
|
|
|
|
return emu;
|
|
|
|
}
|
|
|
|
|
|
|
|
//! Per-channel and per-operator registers map
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
static const uint16_t g_operatorsMap[(NUM_OF_CHANNELS + NUM_OF_RM_CHANNELS) * 2] =
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
|
|
|
// Channels 0-2
|
|
|
|
0x000, 0x003, 0x001, 0x004, 0x002, 0x005, // operators 0, 3, 1, 4, 2, 5
|
|
|
|
// Channels 3-5
|
|
|
|
0x008, 0x00B, 0x009, 0x00C, 0x00A, 0x00D, // operators 6, 9, 7,10, 8,11
|
|
|
|
// Channels 6-8
|
|
|
|
0x010, 0x013, 0x011, 0x014, 0x012, 0x015, // operators 12,15, 13,16, 14,17
|
|
|
|
// Same for second card
|
|
|
|
0x100, 0x103, 0x101, 0x104, 0x102, 0x105, // operators 18,21, 19,22, 20,23
|
|
|
|
0x108, 0x10B, 0x109, 0x10C, 0x10A, 0x10D, // operators 24,27, 25,28, 26,29
|
|
|
|
0x110, 0x113, 0x111, 0x114, 0x112, 0x115, // operators 30,33, 31,34, 32,35
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
|
|
|
|
//==For Rhythm-mode percussions
|
2020-01-02 16:52:30 +00:00
|
|
|
// Channel 18
|
|
|
|
0x010, 0x013, // operators 12,15
|
|
|
|
// Channel 19
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
0xFFF, 0x014, // operator 16
|
|
|
|
// Channel 19
|
|
|
|
0x012, 0xFFF, // operator 14
|
|
|
|
// Channel 19
|
|
|
|
0xFFF, 0x015, // operator 17
|
|
|
|
// Channel 19
|
|
|
|
0x011, 0xFFF, // operator 13
|
|
|
|
|
|
|
|
//==For Rhythm-mode percussions in CMF, snare and cymbal operators has inverted!
|
|
|
|
0x010, 0x013, // operators 12,15
|
|
|
|
// Channel 19
|
2020-01-02 16:52:30 +00:00
|
|
|
0x014, 0xFFF, // operator 16
|
|
|
|
// Channel 19
|
|
|
|
0x012, 0xFFF, // operator 14
|
|
|
|
// Channel 19
|
|
|
|
0x015, 0xFFF, // operator 17
|
|
|
|
// Channel 19
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
0x011, 0xFFF // operator 13
|
|
|
|
};
|
2020-01-02 16:52:30 +00:00
|
|
|
|
|
|
|
//! Channel map to regoster offsets
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
static const uint16_t g_channelsMap[NUM_OF_CHANNELS] =
|
|
|
|
{
|
|
|
|
0x000, 0x001, 0x002, 0x003, 0x004, 0x005, 0x006, 0x007, 0x008, // 0..8
|
|
|
|
0x100, 0x101, 0x102, 0x103, 0x104, 0x105, 0x106, 0x107, 0x108, // 9..17 (secondary set)
|
|
|
|
0x006, 0x007, 0x008, 0x008, 0x008 // <- hw percussions, hihats and cymbals using tom-tom's channel as pitch source
|
|
|
|
};
|
|
|
|
|
|
|
|
//! Channel map to regoster offsets (separated copy for panning)
|
|
|
|
static const uint16_t g_channelsMapPan[NUM_OF_CHANNELS] =
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
|
|
|
0x000, 0x001, 0x002, 0x003, 0x004, 0x005, 0x006, 0x007, 0x008, // 0..8
|
|
|
|
0x100, 0x101, 0x102, 0x103, 0x104, 0x105, 0x106, 0x107, 0x108, // 9..17 (secondary set)
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
0x006, 0x007, 0x008, 0xFFF, 0xFFF // <- hw percussions, 0xFFF = no support for pitch/pan
|
|
|
|
};
|
2020-01-02 16:52:30 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
In OPL3 mode:
|
|
|
|
0 1 2 6 7 8 9 10 11 16 17 18
|
|
|
|
op0 op1 op2 op12 op13 op14 op18 op19 op20 op30 op31 op32
|
|
|
|
op3 op4 op5 op15 op16 op17 op21 op22 op23 op33 op34 op35
|
|
|
|
3 4 5 13 14 15
|
|
|
|
op6 op7 op8 op24 op25 op26
|
|
|
|
op9 op10 op11 op27 op28 op29
|
|
|
|
Ports:
|
|
|
|
+0 +1 +2 +10 +11 +12 +100 +101 +102 +110 +111 +112
|
|
|
|
+3 +4 +5 +13 +14 +15 +103 +104 +105 +113 +114 +115
|
|
|
|
+8 +9 +A +108 +109 +10A
|
|
|
|
+B +C +D +10B +10C +10D
|
|
|
|
|
|
|
|
Percussion:
|
|
|
|
bassdrum = op(0): 0xBD bit 0x10, operators 12 (0x10) and 15 (0x13) / channels 6, 6b
|
|
|
|
snare = op(3): 0xBD bit 0x08, operators 16 (0x14) / channels 7b
|
|
|
|
tomtom = op(4): 0xBD bit 0x04, operators 14 (0x12) / channels 8
|
|
|
|
cym = op(5): 0xBD bit 0x02, operators 17 (0x17) / channels 8b
|
|
|
|
hihat = op(2): 0xBD bit 0x01, operators 13 (0x11) / channels 7
|
|
|
|
|
|
|
|
|
|
|
|
In OPTi mode ("extended FM" in 82C924, 82C925, 82C931 chips):
|
|
|
|
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
|
|
|
|
op0 op4 op6 op10 op12 op16 op18 op22 op24 op28 op30 op34 op36 op38 op40 op42 op44 op46
|
|
|
|
op1 op5 op7 op11 op13 op17 op19 op23 op25 op29 op31 op35 op37 op39 op41 op43 op45 op47
|
|
|
|
op2 op8 op14 op20 op26 op32
|
|
|
|
op3 op9 op15 op21 op27 op33 for a total of 6 quad + 12 dual
|
|
|
|
Ports: ???
|
|
|
|
*/
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/***************************************************************
|
|
|
|
* Volume model tables *
|
|
|
|
***************************************************************/
|
|
|
|
|
|
|
|
// Mapping from MIDI volume level to OPL level value.
|
|
|
|
|
|
|
|
static const uint_fast32_t s_dmx_volume_model[128] =
|
|
|
|
{
|
|
|
|
0, 1, 3, 5, 6, 8, 10, 11,
|
|
|
|
13, 14, 16, 17, 19, 20, 22, 23,
|
|
|
|
25, 26, 27, 29, 30, 32, 33, 34,
|
|
|
|
36, 37, 39, 41, 43, 45, 47, 49,
|
|
|
|
50, 52, 54, 55, 57, 59, 60, 61,
|
|
|
|
63, 64, 66, 67, 68, 69, 71, 72,
|
|
|
|
73, 74, 75, 76, 77, 79, 80, 81,
|
|
|
|
82, 83, 84, 84, 85, 86, 87, 88,
|
|
|
|
89, 90, 91, 92, 92, 93, 94, 95,
|
|
|
|
96, 96, 97, 98, 99, 99, 100, 101,
|
|
|
|
101, 102, 103, 103, 104, 105, 105, 106,
|
|
|
|
107, 107, 108, 109, 109, 110, 110, 111,
|
|
|
|
112, 112, 113, 113, 114, 114, 115, 115,
|
|
|
|
116, 117, 117, 118, 118, 119, 119, 120,
|
|
|
|
120, 121, 121, 122, 122, 123, 123, 123,
|
|
|
|
124, 124, 125, 125, 126, 126, 127, 127,
|
|
|
|
};
|
|
|
|
|
|
|
|
static const uint_fast32_t s_w9x_sb16_volume_model[32] =
|
|
|
|
{
|
|
|
|
80, 63, 40, 36, 32, 28, 23, 21,
|
|
|
|
19, 17, 15, 14, 13, 12, 11, 10,
|
|
|
|
9, 8, 7, 6, 5, 5, 4, 4,
|
|
|
|
3, 3, 2, 2, 1, 1, 0, 0
|
|
|
|
};
|
|
|
|
|
|
|
|
static const uint_fast32_t s_w9x_generic_fm_volume_model[32] =
|
|
|
|
{
|
|
|
|
40, 36, 32, 28, 23, 21, 19, 17,
|
|
|
|
15, 14, 13, 12, 11, 10, 9, 8,
|
|
|
|
7, 6, 5, 5, 4, 4, 3, 3,
|
|
|
|
2, 2, 1, 1, 1, 0, 0, 0
|
|
|
|
};
|
|
|
|
|
|
|
|
static const uint_fast32_t s_ail_vel_graph[16] =
|
|
|
|
{
|
|
|
|
82, 85, 88, 91, 94, 97, 100, 103,
|
|
|
|
106, 109, 112, 115, 118, 121, 124, 127
|
|
|
|
};
|
|
|
|
|
|
|
|
static const uint_fast32_t s_hmi_volume_table[64] =
|
|
|
|
{
|
|
|
|
0x3F, 0x3A, 0x35, 0x30, 0x2C, 0x29, 0x25, 0x24,
|
|
|
|
0x23, 0x22, 0x21, 0x20, 0x1F, 0x1E, 0x1D, 0x1C,
|
|
|
|
0x1B, 0x1A, 0x19, 0x18, 0x17, 0x16, 0x15, 0x14,
|
|
|
|
0x13, 0x12, 0x11, 0x10, 0x0F, 0x0E, 0x0E, 0x0D,
|
|
|
|
0x0D, 0x0C, 0x0C, 0x0B, 0x0B, 0x0A, 0x0A, 0x09,
|
|
|
|
0x09, 0x08, 0x08, 0x07, 0x07, 0x06, 0x06, 0x06,
|
|
|
|
0x05, 0x05, 0x05, 0x04, 0x04, 0x04, 0x04, 0x03,
|
|
|
|
0x03, 0x03, 0x02, 0x02, 0x02, 0x01, 0x01, 0x00,
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/***************************************************************
|
|
|
|
* Standard frequency formula *
|
|
|
|
* *************************************************************/
|
|
|
|
|
|
|
|
static inline double s_commonFreq(double tone)
|
|
|
|
{
|
|
|
|
return BEND_COEFFICIENT * std::exp(0.057762265 * tone);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/***************************************************************
|
|
|
|
* DMX frequency model *
|
|
|
|
* *************************************************************/
|
|
|
|
|
|
|
|
// DMX volumes table
|
|
|
|
static const int_fast32_t s_dmx_freq_table[] =
|
|
|
|
{
|
|
|
|
0x0133, 0x0133, 0x0134, 0x0134, 0x0135, 0x0136, 0x0136, 0x0137,
|
|
|
|
0x0137, 0x0138, 0x0138, 0x0139, 0x0139, 0x013A, 0x013B, 0x013B,
|
|
|
|
0x013C, 0x013C, 0x013D, 0x013D, 0x013E, 0x013F, 0x013F, 0x0140,
|
|
|
|
0x0140, 0x0141, 0x0142, 0x0142, 0x0143, 0x0143, 0x0144, 0x0144,
|
|
|
|
|
|
|
|
0x0145, 0x0146, 0x0146, 0x0147, 0x0147, 0x0148, 0x0149, 0x0149,
|
|
|
|
0x014A, 0x014A, 0x014B, 0x014C, 0x014C, 0x014D, 0x014D, 0x014E,
|
|
|
|
0x014F, 0x014F, 0x0150, 0x0150, 0x0151, 0x0152, 0x0152, 0x0153,
|
|
|
|
0x0153, 0x0154, 0x0155, 0x0155, 0x0156, 0x0157, 0x0157, 0x0158,
|
|
|
|
|
|
|
|
0x0158, 0x0159, 0x015A, 0x015A, 0x015B, 0x015B, 0x015C, 0x015D,
|
|
|
|
0x015D, 0x015E, 0x015F, 0x015F, 0x0160, 0x0161, 0x0161, 0x0162,
|
|
|
|
0x0162, 0x0163, 0x0164, 0x0164, 0x0165, 0x0166, 0x0166, 0x0167,
|
|
|
|
0x0168, 0x0168, 0x0169, 0x016A, 0x016A, 0x016B, 0x016C, 0x016C,
|
|
|
|
|
|
|
|
0x016D, 0x016E, 0x016E, 0x016F, 0x0170, 0x0170, 0x0171, 0x0172,
|
|
|
|
0x0172, 0x0173, 0x0174, 0x0174, 0x0175, 0x0176, 0x0176, 0x0177,
|
|
|
|
0x0178, 0x0178, 0x0179, 0x017A, 0x017A, 0x017B, 0x017C, 0x017C,
|
|
|
|
0x017D, 0x017E, 0x017E, 0x017F, 0x0180, 0x0181, 0x0181, 0x0182,
|
|
|
|
|
|
|
|
0x0183, 0x0183, 0x0184, 0x0185, 0x0185, 0x0186, 0x0187, 0x0188,
|
|
|
|
0x0188, 0x0189, 0x018A, 0x018A, 0x018B, 0x018C, 0x018D, 0x018D,
|
|
|
|
0x018E, 0x018F, 0x018F, 0x0190, 0x0191, 0x0192, 0x0192, 0x0193,
|
|
|
|
0x0194, 0x0194, 0x0195, 0x0196, 0x0197, 0x0197, 0x0198, 0x0199,
|
|
|
|
|
|
|
|
0x019A, 0x019A, 0x019B, 0x019C, 0x019D, 0x019D, 0x019E, 0x019F,
|
|
|
|
0x01A0, 0x01A0, 0x01A1, 0x01A2, 0x01A3, 0x01A3, 0x01A4, 0x01A5,
|
|
|
|
0x01A6, 0x01A6, 0x01A7, 0x01A8, 0x01A9, 0x01A9, 0x01AA, 0x01AB,
|
|
|
|
0x01AC, 0x01AD, 0x01AD, 0x01AE, 0x01AF, 0x01B0, 0x01B0, 0x01B1,
|
|
|
|
|
|
|
|
0x01B2, 0x01B3, 0x01B4, 0x01B4, 0x01B5, 0x01B6, 0x01B7, 0x01B8,
|
|
|
|
0x01B8, 0x01B9, 0x01BA, 0x01BB, 0x01BC, 0x01BC, 0x01BD, 0x01BE,
|
|
|
|
0x01BF, 0x01C0, 0x01C0, 0x01C1, 0x01C2, 0x01C3, 0x01C4, 0x01C4,
|
|
|
|
0x01C5, 0x01C6, 0x01C7, 0x01C8, 0x01C9, 0x01C9, 0x01CA, 0x01CB,
|
|
|
|
|
|
|
|
0x01CC, 0x01CD, 0x01CE, 0x01CE, 0x01CF, 0x01D0, 0x01D1, 0x01D2,
|
|
|
|
0x01D3, 0x01D3, 0x01D4, 0x01D5, 0x01D6, 0x01D7, 0x01D8, 0x01D8,
|
|
|
|
0x01D9, 0x01DA, 0x01DB, 0x01DC, 0x01DD, 0x01DE, 0x01DE, 0x01DF,
|
|
|
|
0x01E0, 0x01E1, 0x01E2, 0x01E3, 0x01E4, 0x01E5, 0x01E5, 0x01E6,
|
|
|
|
|
|
|
|
0x01E7, 0x01E8, 0x01E9, 0x01EA, 0x01EB, 0x01EC, 0x01ED, 0x01ED,
|
|
|
|
0x01EE, 0x01EF, 0x01F0, 0x01F1, 0x01F2, 0x01F3, 0x01F4, 0x01F5,
|
|
|
|
0x01F6, 0x01F6, 0x01F7, 0x01F8, 0x01F9, 0x01FA, 0x01FB, 0x01FC,
|
|
|
|
0x01FD, 0x01FE, 0x01FF,
|
|
|
|
|
|
|
|
0x0200, 0x0201, 0x0201, 0x0202, 0x0203, 0x0204, 0x0205, 0x0206,
|
|
|
|
0x0207, 0x0208, 0x0209, 0x020A, 0x020B, 0x020C, 0x020D, 0x020E,
|
|
|
|
0x020F, 0x0210, 0x0210, 0x0211, 0x0212, 0x0213, 0x0214, 0x0215,
|
|
|
|
0x0216, 0x0217, 0x0218, 0x0219, 0x021A, 0x021B, 0x021C, 0x021D,
|
|
|
|
|
|
|
|
0x021E, 0x021F, 0x0220, 0x0221, 0x0222, 0x0223, 0x0224, 0x0225,
|
|
|
|
0x0226, 0x0227, 0x0228, 0x0229, 0x022A, 0x022B, 0x022C, 0x022D,
|
|
|
|
0x022E, 0x022F, 0x0230, 0x0231, 0x0232, 0x0233, 0x0234, 0x0235,
|
|
|
|
0x0236, 0x0237, 0x0238, 0x0239, 0x023A, 0x023B, 0x023C, 0x023D,
|
|
|
|
|
|
|
|
0x023E, 0x023F, 0x0240, 0x0241, 0x0242, 0x0244, 0x0245, 0x0246,
|
|
|
|
0x0247, 0x0248, 0x0249, 0x024A, 0x024B, 0x024C, 0x024D, 0x024E,
|
|
|
|
0x024F, 0x0250, 0x0251, 0x0252, 0x0253, 0x0254, 0x0256, 0x0257,
|
|
|
|
0x0258, 0x0259, 0x025A, 0x025B, 0x025C, 0x025D, 0x025E, 0x025F,
|
|
|
|
|
|
|
|
0x0260, 0x0262, 0x0263, 0x0264, 0x0265, 0x0266, 0x0267, 0x0268,
|
|
|
|
0x0269, 0x026A, 0x026C, 0x026D, 0x026E, 0x026F, 0x0270, 0x0271,
|
|
|
|
0x0272, 0x0273, 0x0275, 0x0276, 0x0277, 0x0278, 0x0279, 0x027A,
|
|
|
|
0x027B, 0x027D, 0x027E, 0x027F, 0x0280, 0x0281, 0x0282, 0x0284,
|
|
|
|
|
|
|
|
0x0285, 0x0286, 0x0287, 0x0288, 0x0289, 0x028B, 0x028C, 0x028D,
|
|
|
|
0x028E, 0x028F, 0x0290, 0x0292, 0x0293, 0x0294, 0x0295, 0x0296,
|
|
|
|
0x0298, 0x0299, 0x029A, 0x029B, 0x029C, 0x029E, 0x029F, 0x02A0,
|
|
|
|
0x02A1, 0x02A2, 0x02A4, 0x02A5, 0x02A6, 0x02A7, 0x02A9, 0x02AA,
|
|
|
|
|
|
|
|
0x02AB, 0x02AC, 0x02AE, 0x02AF, 0x02B0, 0x02B1, 0x02B2, 0x02B4,
|
|
|
|
0x02B5, 0x02B6, 0x02B7, 0x02B9, 0x02BA, 0x02BB, 0x02BD, 0x02BE,
|
|
|
|
0x02BF, 0x02C0, 0x02C2, 0x02C3, 0x02C4, 0x02C5, 0x02C7, 0x02C8,
|
|
|
|
0x02C9, 0x02CB, 0x02CC, 0x02CD, 0x02CE, 0x02D0, 0x02D1, 0x02D2,
|
|
|
|
|
|
|
|
0x02D4, 0x02D5, 0x02D6, 0x02D8, 0x02D9, 0x02DA, 0x02DC, 0x02DD,
|
|
|
|
0x02DE, 0x02E0, 0x02E1, 0x02E2, 0x02E4, 0x02E5, 0x02E6, 0x02E8,
|
|
|
|
0x02E9, 0x02EA, 0x02EC, 0x02ED, 0x02EE, 0x02F0, 0x02F1, 0x02F2,
|
|
|
|
0x02F4, 0x02F5, 0x02F6, 0x02F8, 0x02F9, 0x02FB, 0x02FC, 0x02FD,
|
|
|
|
|
|
|
|
0x02FF, 0x0300, 0x0302, 0x0303, 0x0304, 0x0306, 0x0307, 0x0309,
|
|
|
|
0x030A, 0x030B, 0x030D, 0x030E, 0x0310, 0x0311, 0x0312, 0x0314,
|
|
|
|
0x0315, 0x0317, 0x0318, 0x031A, 0x031B, 0x031C, 0x031E, 0x031F,
|
|
|
|
0x0321, 0x0322, 0x0324, 0x0325, 0x0327, 0x0328, 0x0329, 0x032B,
|
|
|
|
|
|
|
|
0x032C, 0x032E, 0x032F, 0x0331, 0x0332, 0x0334, 0x0335, 0x0337,
|
|
|
|
0x0338, 0x033A, 0x033B, 0x033D, 0x033E, 0x0340, 0x0341, 0x0343,
|
|
|
|
0x0344, 0x0346, 0x0347, 0x0349, 0x034A, 0x034C, 0x034D, 0x034F,
|
|
|
|
0x0350, 0x0352, 0x0353, 0x0355, 0x0357, 0x0358, 0x035A, 0x035B,
|
|
|
|
|
|
|
|
0x035D, 0x035E, 0x0360, 0x0361, 0x0363, 0x0365, 0x0366, 0x0368,
|
|
|
|
0x0369, 0x036B, 0x036C, 0x036E, 0x0370, 0x0371, 0x0373, 0x0374,
|
|
|
|
0x0376, 0x0378, 0x0379, 0x037B, 0x037C, 0x037E, 0x0380, 0x0381,
|
|
|
|
0x0383, 0x0384, 0x0386, 0x0388, 0x0389, 0x038B, 0x038D, 0x038E,
|
|
|
|
|
|
|
|
0x0390, 0x0392, 0x0393, 0x0395, 0x0397, 0x0398, 0x039A, 0x039C,
|
|
|
|
0x039D, 0x039F, 0x03A1, 0x03A2, 0x03A4, 0x03A6, 0x03A7, 0x03A9,
|
|
|
|
0x03AB, 0x03AC, 0x03AE, 0x03B0, 0x03B1, 0x03B3, 0x03B5, 0x03B7,
|
|
|
|
0x03B8, 0x03BA, 0x03BC, 0x03BD, 0x03BF, 0x03C1, 0x03C3, 0x03C4,
|
|
|
|
|
|
|
|
0x03C6, 0x03C8, 0x03CA, 0x03CB, 0x03CD, 0x03CF, 0x03D1, 0x03D2,
|
|
|
|
0x03D4, 0x03D6, 0x03D8, 0x03DA, 0x03DB, 0x03DD, 0x03DF, 0x03E1,
|
|
|
|
0x03E3, 0x03E4, 0x03E6, 0x03E8, 0x03EA, 0x03EC, 0x03ED, 0x03EF,
|
|
|
|
0x03F1, 0x03F3, 0x03F5, 0x03F6, 0x03F8, 0x03FA, 0x03FC, 0x03FE,
|
|
|
|
|
|
|
|
0x036C
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline double s_dmxFreq(double tone)
|
|
|
|
{
|
|
|
|
uint_fast32_t noteI = (uint_fast32_t)(tone);
|
|
|
|
int_fast32_t bendI = 0;
|
|
|
|
int_fast32_t outHz = 0.0;
|
|
|
|
double bendDec = tone - (int)tone;
|
|
|
|
|
|
|
|
bendI = (int_fast32_t)((bendDec * 128.0) / 2.0) + 128;
|
|
|
|
bendI = bendI >> 1;
|
|
|
|
|
|
|
|
int_fast32_t oct = 0;
|
|
|
|
int_fast32_t freqIndex = (noteI << 5) + bendI;
|
|
|
|
|
|
|
|
if(freqIndex < 0)
|
|
|
|
freqIndex = 0;
|
|
|
|
else if(freqIndex >= 284)
|
|
|
|
{
|
|
|
|
freqIndex -= 284;
|
|
|
|
oct = freqIndex / 384;
|
|
|
|
freqIndex = (freqIndex % 384) + 284;
|
|
|
|
}
|
|
|
|
|
|
|
|
outHz = s_dmx_freq_table[freqIndex];
|
|
|
|
|
|
|
|
while(oct > 1)
|
|
|
|
{
|
|
|
|
outHz *= 2;
|
|
|
|
oct -= 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return (double)outHz;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/***************************************************************
|
|
|
|
* Apogee Sound System frequency model *
|
|
|
|
***************************************************************/
|
|
|
|
|
|
|
|
static const int_fast32_t s_apogee_freq_table[31 + 1][12] =
|
|
|
|
{
|
|
|
|
{ 0x157, 0x16b, 0x181, 0x198, 0x1b0, 0x1ca, 0x1e5, 0x202, 0x220, 0x241, 0x263, 0x287 },
|
|
|
|
{ 0x157, 0x16b, 0x181, 0x198, 0x1b0, 0x1ca, 0x1e5, 0x202, 0x220, 0x242, 0x264, 0x288 },
|
|
|
|
{ 0x158, 0x16c, 0x182, 0x199, 0x1b1, 0x1cb, 0x1e6, 0x203, 0x221, 0x243, 0x265, 0x289 },
|
|
|
|
{ 0x158, 0x16c, 0x183, 0x19a, 0x1b2, 0x1cc, 0x1e7, 0x204, 0x222, 0x244, 0x266, 0x28a },
|
|
|
|
{ 0x159, 0x16d, 0x183, 0x19a, 0x1b3, 0x1cd, 0x1e8, 0x205, 0x223, 0x245, 0x267, 0x28b },
|
|
|
|
{ 0x15a, 0x16e, 0x184, 0x19b, 0x1b3, 0x1ce, 0x1e9, 0x206, 0x224, 0x246, 0x268, 0x28c },
|
|
|
|
{ 0x15a, 0x16e, 0x185, 0x19c, 0x1b4, 0x1ce, 0x1ea, 0x207, 0x225, 0x247, 0x269, 0x28e },
|
|
|
|
{ 0x15b, 0x16f, 0x185, 0x19d, 0x1b5, 0x1cf, 0x1eb, 0x208, 0x226, 0x248, 0x26a, 0x28f },
|
|
|
|
{ 0x15b, 0x170, 0x186, 0x19d, 0x1b6, 0x1d0, 0x1ec, 0x209, 0x227, 0x249, 0x26b, 0x290 },
|
|
|
|
{ 0x15c, 0x170, 0x187, 0x19e, 0x1b7, 0x1d1, 0x1ec, 0x20a, 0x228, 0x24a, 0x26d, 0x291 },
|
|
|
|
{ 0x15d, 0x171, 0x188, 0x19f, 0x1b7, 0x1d2, 0x1ed, 0x20b, 0x229, 0x24b, 0x26e, 0x292 },
|
|
|
|
{ 0x15d, 0x172, 0x188, 0x1a0, 0x1b8, 0x1d3, 0x1ee, 0x20c, 0x22a, 0x24c, 0x26f, 0x293 },
|
|
|
|
{ 0x15e, 0x172, 0x189, 0x1a0, 0x1b9, 0x1d4, 0x1ef, 0x20d, 0x22b, 0x24d, 0x270, 0x295 },
|
|
|
|
{ 0x15f, 0x173, 0x18a, 0x1a1, 0x1ba, 0x1d4, 0x1f0, 0x20e, 0x22c, 0x24e, 0x271, 0x296 },
|
|
|
|
{ 0x15f, 0x174, 0x18a, 0x1a2, 0x1bb, 0x1d5, 0x1f1, 0x20f, 0x22d, 0x24f, 0x272, 0x297 },
|
|
|
|
{ 0x160, 0x174, 0x18b, 0x1a3, 0x1bb, 0x1d6, 0x1f2, 0x210, 0x22e, 0x250, 0x273, 0x298 },
|
|
|
|
{ 0x161, 0x175, 0x18c, 0x1a3, 0x1bc, 0x1d7, 0x1f3, 0x211, 0x22f, 0x251, 0x274, 0x299 },
|
|
|
|
{ 0x161, 0x176, 0x18c, 0x1a4, 0x1bd, 0x1d8, 0x1f4, 0x212, 0x230, 0x252, 0x276, 0x29b },
|
|
|
|
{ 0x162, 0x176, 0x18d, 0x1a5, 0x1be, 0x1d9, 0x1f5, 0x212, 0x231, 0x254, 0x277, 0x29c },
|
|
|
|
{ 0x162, 0x177, 0x18e, 0x1a6, 0x1bf, 0x1d9, 0x1f5, 0x213, 0x232, 0x255, 0x278, 0x29d },
|
|
|
|
{ 0x163, 0x178, 0x18f, 0x1a6, 0x1bf, 0x1da, 0x1f6, 0x214, 0x233, 0x256, 0x279, 0x29e },
|
|
|
|
{ 0x164, 0x179, 0x18f, 0x1a7, 0x1c0, 0x1db, 0x1f7, 0x215, 0x235, 0x257, 0x27a, 0x29f },
|
|
|
|
{ 0x164, 0x179, 0x190, 0x1a8, 0x1c1, 0x1dc, 0x1f8, 0x216, 0x236, 0x258, 0x27b, 0x2a1 },
|
|
|
|
{ 0x165, 0x17a, 0x191, 0x1a9, 0x1c2, 0x1dd, 0x1f9, 0x217, 0x237, 0x259, 0x27c, 0x2a2 },
|
|
|
|
{ 0x166, 0x17b, 0x192, 0x1aa, 0x1c3, 0x1de, 0x1fa, 0x218, 0x238, 0x25a, 0x27e, 0x2a3 },
|
|
|
|
{ 0x166, 0x17b, 0x192, 0x1aa, 0x1c3, 0x1df, 0x1fb, 0x219, 0x239, 0x25b, 0x27f, 0x2a4 },
|
|
|
|
{ 0x167, 0x17c, 0x193, 0x1ab, 0x1c4, 0x1e0, 0x1fc, 0x21a, 0x23a, 0x25c, 0x280, 0x2a6 },
|
|
|
|
{ 0x168, 0x17d, 0x194, 0x1ac, 0x1c5, 0x1e0, 0x1fd, 0x21b, 0x23b, 0x25d, 0x281, 0x2a7 },
|
|
|
|
{ 0x168, 0x17d, 0x194, 0x1ad, 0x1c6, 0x1e1, 0x1fe, 0x21c, 0x23c, 0x25e, 0x282, 0x2a8 },
|
|
|
|
{ 0x169, 0x17e, 0x195, 0x1ad, 0x1c7, 0x1e2, 0x1ff, 0x21d, 0x23d, 0x260, 0x283, 0x2a9 },
|
|
|
|
{ 0x16a, 0x17f, 0x196, 0x1ae, 0x1c8, 0x1e3, 0x1ff, 0x21e, 0x23e, 0x261, 0x284, 0x2ab },
|
|
|
|
{ 0x16a, 0x17f, 0x197, 0x1af, 0x1c8, 0x1e4, 0x200, 0x21f, 0x23f, 0x262, 0x286, 0x2ac }
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline double s_apogeeFreq(double tone)
|
|
|
|
{
|
|
|
|
uint_fast32_t noteI = (uint_fast32_t)(tone);
|
|
|
|
int_fast32_t bendI = 0;
|
|
|
|
int_fast32_t outHz = 0.0;
|
|
|
|
double bendDec = tone - (int)tone;
|
|
|
|
int_fast32_t octave;
|
|
|
|
int_fast32_t scaleNote;
|
|
|
|
|
|
|
|
bendI = (int_fast32_t)(bendDec * 32) + 32;
|
|
|
|
|
|
|
|
noteI += bendI / 32;
|
|
|
|
noteI -= 1;
|
|
|
|
|
|
|
|
scaleNote = noteI % 12;
|
|
|
|
octave = noteI / 12;
|
|
|
|
|
|
|
|
outHz = s_apogee_freq_table[bendI % 32][scaleNote];
|
|
|
|
|
|
|
|
while(octave > 1)
|
|
|
|
{
|
|
|
|
outHz *= 2;
|
|
|
|
octave -= 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return (double)outHz;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/***************************************************************
|
|
|
|
* Windows 9x FM drivers frequency model *
|
|
|
|
***************************************************************/
|
|
|
|
|
|
|
|
//static const double s_9x_opl_samplerate = 50000.0;
|
|
|
|
//static const double s_9x_opl_tune = 440.0;
|
|
|
|
static const uint_fast8_t s_9x_opl_pitchfrac = 8;
|
|
|
|
|
|
|
|
static const uint_fast32_t s_9x_opl_freq[12] =
|
|
|
|
{
|
|
|
|
0xAB7, 0xB5A, 0xC07, 0xCBE, 0xD80, 0xE4D, 0xF27, 0x100E, 0x1102, 0x1205, 0x1318, 0x143A
|
|
|
|
};
|
|
|
|
|
|
|
|
static const int32_t s_9x_opl_uppitch = 31;
|
|
|
|
static const int32_t s_9x_opl_downpitch = 27;
|
|
|
|
|
|
|
|
static uint_fast32_t s_9x_opl_applypitch(uint_fast32_t freq, int_fast32_t pitch)
|
|
|
|
{
|
|
|
|
int32_t diff;
|
|
|
|
|
|
|
|
if(pitch > 0)
|
|
|
|
{
|
|
|
|
diff = (pitch * s_9x_opl_uppitch) >> s_9x_opl_pitchfrac;
|
|
|
|
freq += (diff * freq) >> 15;
|
|
|
|
}
|
|
|
|
else if (pitch < 0)
|
|
|
|
{
|
|
|
|
diff = (-pitch * s_9x_opl_downpitch) >> s_9x_opl_pitchfrac;
|
|
|
|
freq -= (diff * freq) >> 15;
|
|
|
|
}
|
|
|
|
|
|
|
|
return freq;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline double s_9xFreq(double tone)
|
|
|
|
{
|
|
|
|
uint_fast32_t note = (uint_fast32_t)(tone);
|
|
|
|
int_fast32_t bend;
|
|
|
|
double bendDec = tone - (int)tone; // 0.0 ± 1.0 - one halftone
|
|
|
|
|
|
|
|
uint_fast32_t freq;
|
|
|
|
uint_fast32_t freqpitched;
|
|
|
|
uint_fast32_t octave;
|
|
|
|
|
|
|
|
uint_fast32_t bendMsb;
|
|
|
|
uint_fast32_t bendLsb;
|
|
|
|
|
|
|
|
bend = (int_fast32_t)(bendDec * 4096) + 8192; // convert to MIDI standard value
|
|
|
|
|
|
|
|
bendMsb = (bend >> 7) & 0x7F;
|
|
|
|
bendLsb = (bend & 0x7F);
|
|
|
|
|
|
|
|
bend = (bendMsb << 9) | (bendLsb << 2);
|
|
|
|
bend = (int16_t)(uint16_t)(bend + 0x8000);
|
|
|
|
|
|
|
|
octave = note / 12;
|
|
|
|
freq = s_9x_opl_freq[note % 12];
|
|
|
|
if(octave < 5)
|
|
|
|
freq >>= (5 - octave);
|
|
|
|
else if (octave > 5)
|
|
|
|
freq <<= (octave - 5);
|
|
|
|
|
|
|
|
freqpitched = s_9x_opl_applypitch(freq, bend);
|
|
|
|
freqpitched *= 2;
|
|
|
|
|
|
|
|
return (double)freqpitched;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/***************************************************************
|
|
|
|
* HMI Sound Operating System frequency model *
|
|
|
|
***************************************************************/
|
|
|
|
|
|
|
|
const size_t s_hmi_freqtable_size = 103;
|
|
|
|
static uint_fast32_t s_hmi_freqtable[s_hmi_freqtable_size] =
|
|
|
|
{
|
|
|
|
0x0157, 0x016B, 0x0181, 0x0198, 0x01B0, 0x01CA, 0x01E5, 0x0202, 0x0220, 0x0241, 0x0263, 0x0287,
|
|
|
|
0x0557, 0x056B, 0x0581, 0x0598, 0x05B0, 0x05CA, 0x05E5, 0x0602, 0x0620, 0x0641, 0x0663, 0x0687,
|
|
|
|
0x0957, 0x096B, 0x0981, 0x0998, 0x09B0, 0x09CA, 0x09E5, 0x0A02, 0x0A20, 0x0A41, 0x0A63, 0x0A87,
|
|
|
|
0x0D57, 0x0D6B, 0x0D81, 0x0D98, 0x0DB0, 0x0DCA, 0x0DE5, 0x0E02, 0x0E20, 0x0E41, 0x0E63, 0x0E87,
|
|
|
|
0x1157, 0x116B, 0x1181, 0x1198, 0x11B0, 0x11CA, 0x11E5, 0x1202, 0x1220, 0x1241, 0x1263, 0x1287,
|
|
|
|
0x1557, 0x156B, 0x1581, 0x1598, 0x15B0, 0x15CA, 0x15E5, 0x1602, 0x1620, 0x1641, 0x1663, 0x1687,
|
|
|
|
0x1957, 0x196B, 0x1981, 0x1998, 0x19B0, 0x19CA, 0x19E5, 0x1A02, 0x1A20, 0x1A41, 0x1A63, 0x1A87,
|
|
|
|
0x1D57, 0x1D6B, 0x1D81, 0x1D98, 0x1DB0, 0x1DCA, 0x1DE5, 0x1E02, 0x1E20, 0x1E41, 0x1E63, 0x1E87,
|
|
|
|
0x1EAE, 0x1EB7, 0x1F02, 0x1F30, 0x1F60, 0x1F94, 0x1FCA
|
|
|
|
};
|
|
|
|
|
|
|
|
const size_t s_hmi_bendtable_size = 12;
|
|
|
|
static uint_fast32_t s_hmi_bendtable[s_hmi_bendtable_size] =
|
|
|
|
{
|
|
|
|
0x144, 0x132, 0x121, 0x110, 0x101, 0xf8, 0xe5, 0xd8, 0xcc, 0xc1, 0xb6, 0xac
|
|
|
|
};
|
|
|
|
|
|
|
|
#define hmi_range_fix(formula, maxVal) \
|
|
|
|
( \
|
|
|
|
(formula) < 0 ? \
|
|
|
|
0 : \
|
|
|
|
( \
|
|
|
|
(formula) >= (int32_t)maxVal ? \
|
|
|
|
(int32_t)maxVal : \
|
|
|
|
(formula) \
|
|
|
|
)\
|
|
|
|
)
|
|
|
|
|
|
|
|
static uint_fast32_t s_hmi_bend_calc(uint_fast32_t bend, int_fast32_t note)
|
|
|
|
{
|
|
|
|
const int_fast32_t midi_bend_range = 1;
|
|
|
|
uint_fast32_t bendFactor, outFreq, fmOctave, fmFreq, newFreq, idx;
|
|
|
|
int_fast32_t noteMod12;
|
|
|
|
|
|
|
|
note -= 12;
|
|
|
|
// while(doNote >= 12) // ugly way to MOD 12
|
|
|
|
// doNote -= 12;
|
|
|
|
noteMod12 = (note % 12);
|
|
|
|
|
|
|
|
outFreq = s_hmi_freqtable[note];
|
|
|
|
|
|
|
|
fmOctave = outFreq & 0x1c00;
|
|
|
|
fmFreq = outFreq & 0x3ff;
|
|
|
|
|
|
|
|
if(bend < 64)
|
|
|
|
{
|
|
|
|
bendFactor = ((63 - bend) * 1000) >> 6;
|
|
|
|
|
|
|
|
idx = hmi_range_fix(note - midi_bend_range, s_hmi_freqtable_size);
|
|
|
|
newFreq = outFreq - s_hmi_freqtable[idx];
|
|
|
|
|
|
|
|
if(newFreq > 719)
|
|
|
|
{
|
|
|
|
newFreq = fmFreq - s_hmi_bendtable[midi_bend_range - 1];
|
|
|
|
newFreq &= 0x3ff;
|
|
|
|
}
|
|
|
|
|
|
|
|
newFreq = (newFreq * bendFactor) / 1000;
|
|
|
|
outFreq -= newFreq;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
bendFactor = ((bend - 64) * 1000) >> 6;
|
|
|
|
|
|
|
|
idx = hmi_range_fix(note + midi_bend_range, s_hmi_freqtable_size);
|
|
|
|
newFreq = s_hmi_freqtable[idx] - outFreq;
|
|
|
|
|
|
|
|
if(newFreq > 719)
|
|
|
|
{
|
|
|
|
idx = hmi_range_fix(11 - noteMod12, s_hmi_bendtable_size);
|
|
|
|
fmFreq = s_hmi_bendtable[idx];
|
|
|
|
outFreq = (fmOctave + 1024) | fmFreq;
|
|
|
|
|
|
|
|
idx = hmi_range_fix(note + midi_bend_range, s_hmi_freqtable_size);
|
|
|
|
newFreq = s_hmi_freqtable[idx] - outFreq;
|
|
|
|
}
|
|
|
|
|
|
|
|
newFreq = (newFreq * bendFactor) / 1000;
|
|
|
|
outFreq += newFreq;
|
|
|
|
}
|
|
|
|
|
|
|
|
return outFreq;
|
|
|
|
}
|
|
|
|
#undef hmi_range_fix
|
|
|
|
|
|
|
|
static inline double s_hmiFreq(double tone)
|
|
|
|
{
|
|
|
|
int_fast32_t note = (int_fast32_t)(tone);
|
|
|
|
double bendDec = tone - (int)tone; // 0.0 ± 1.0 - one halftone
|
|
|
|
int_fast32_t bend;
|
|
|
|
uint_fast32_t inFreq;
|
|
|
|
uint_fast32_t freq;
|
|
|
|
int_fast32_t octave;
|
|
|
|
int_fast32_t octaveOffset = 0;
|
|
|
|
|
|
|
|
bend = (int_fast32_t)(bendDec * 64.0) + 64;
|
|
|
|
|
|
|
|
while(note < 12)
|
|
|
|
{
|
|
|
|
octaveOffset--;
|
|
|
|
note += 12;
|
|
|
|
}
|
|
|
|
while(note > 114)
|
|
|
|
{
|
|
|
|
octaveOffset++;
|
|
|
|
note -= 12;
|
|
|
|
}
|
|
|
|
|
|
|
|
if(bend == 64)
|
|
|
|
inFreq = s_hmi_freqtable[note - 12];
|
|
|
|
else
|
|
|
|
inFreq = s_hmi_bend_calc(bend, note);
|
|
|
|
|
|
|
|
freq = inFreq & 0x3FF;
|
|
|
|
octave = (inFreq >> 10) & 0x07;
|
|
|
|
|
|
|
|
octave += octaveOffset;
|
|
|
|
|
|
|
|
while(octave > 0)
|
|
|
|
{
|
|
|
|
freq *= 2;
|
|
|
|
octave -= 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return freq;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/***************************************************************
|
|
|
|
* Audio Interface Library frequency model *
|
|
|
|
***************************************************************/
|
|
|
|
|
|
|
|
static const uint_fast16_t s_ail_freqtable[] = {
|
|
|
|
0x02b2, 0x02b4, 0x02b7, 0x02b9, 0x02bc, 0x02be, 0x02c1, 0x02c3,
|
|
|
|
0x02c6, 0x02c9, 0x02cb, 0x02ce, 0x02d0, 0x02d3, 0x02d6, 0x02d8,
|
|
|
|
0x02db, 0x02dd, 0x02e0, 0x02e3, 0x02e5, 0x02e8, 0x02eb, 0x02ed,
|
|
|
|
0x02f0, 0x02f3, 0x02f6, 0x02f8, 0x02fb, 0x02fe, 0x0301, 0x0303,
|
|
|
|
0x0306, 0x0309, 0x030c, 0x030f, 0x0311, 0x0314, 0x0317, 0x031a,
|
|
|
|
0x031d, 0x0320, 0x0323, 0x0326, 0x0329, 0x032b, 0x032e, 0x0331,
|
|
|
|
0x0334, 0x0337, 0x033a, 0x033d, 0x0340, 0x0343, 0x0346, 0x0349,
|
|
|
|
0x034c, 0x034f, 0x0352, 0x0356, 0x0359, 0x035c, 0x035f, 0x0362,
|
|
|
|
0x0365, 0x0368, 0x036b, 0x036f, 0x0372, 0x0375, 0x0378, 0x037b,
|
|
|
|
0x037f, 0x0382, 0x0385, 0x0388, 0x038c, 0x038f, 0x0392, 0x0395,
|
|
|
|
0x0399, 0x039c, 0x039f, 0x03a3, 0x03a6, 0x03a9, 0x03ad, 0x03b0,
|
|
|
|
0x03b4, 0x03b7, 0x03bb, 0x03be, 0x03c1, 0x03c5, 0x03c8, 0x03cc,
|
|
|
|
0x03cf, 0x03d3, 0x03d7, 0x03da, 0x03de, 0x03e1, 0x03e5, 0x03e8,
|
|
|
|
0x03ec, 0x03f0, 0x03f3, 0x03f7, 0x03fb, 0x03fe, 0xfe01, 0xfe03,
|
|
|
|
0xfe05, 0xfe07, 0xfe08, 0xfe0a, 0xfe0c, 0xfe0e, 0xfe10, 0xfe12,
|
|
|
|
0xfe14, 0xfe16, 0xfe18, 0xfe1a, 0xfe1c, 0xfe1e, 0xfe20, 0xfe21,
|
|
|
|
0xfe23, 0xfe25, 0xfe27, 0xfe29, 0xfe2b, 0xfe2d, 0xfe2f, 0xfe31,
|
|
|
|
0xfe34, 0xfe36, 0xfe38, 0xfe3a, 0xfe3c, 0xfe3e, 0xfe40, 0xfe42,
|
|
|
|
0xfe44, 0xfe46, 0xfe48, 0xfe4a, 0xfe4c, 0xfe4f, 0xfe51, 0xfe53,
|
|
|
|
0xfe55, 0xfe57, 0xfe59, 0xfe5c, 0xfe5e, 0xfe60, 0xfe62, 0xfe64,
|
|
|
|
0xfe67, 0xfe69, 0xfe6b, 0xfe6d, 0xfe6f, 0xfe72, 0xfe74, 0xfe76,
|
|
|
|
0xfe79, 0xfe7b, 0xfe7d, 0xfe7f, 0xfe82, 0xfe84, 0xfe86, 0xfe89,
|
|
|
|
0xfe8b, 0xfe8d, 0xfe90, 0xfe92, 0xfe95, 0xfe97, 0xfe99, 0xfe9c,
|
|
|
|
0xfe9e, 0xfea1, 0xfea3, 0xfea5, 0xfea8, 0xfeaa, 0xfead, 0xfeaf
|
|
|
|
};
|
|
|
|
|
|
|
|
static const uint_fast8_t s_ail_note_octave[] = {
|
|
|
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
|
|
|
|
0x00, 0x00, 0x00, 0x00, 0x01, 0x01, 0x01, 0x01,
|
|
|
|
0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
|
|
|
|
0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02,
|
|
|
|
0x02, 0x02, 0x02, 0x02, 0x03, 0x03, 0x03, 0x03,
|
|
|
|
0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03,
|
|
|
|
0x04, 0x04, 0x04, 0x04, 0x04, 0x04, 0x04, 0x04,
|
|
|
|
0x04, 0x04, 0x04, 0x04, 0x05, 0x05, 0x05, 0x05,
|
|
|
|
0x05, 0x05, 0x05, 0x05, 0x05, 0x05, 0x05, 0x05,
|
|
|
|
0x06, 0x06, 0x06, 0x06, 0x06, 0x06, 0x06, 0x06,
|
|
|
|
0x06, 0x06, 0x06, 0x06, 0x07, 0x07, 0x07, 0x07,
|
|
|
|
0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07
|
|
|
|
};
|
|
|
|
|
|
|
|
static const uint_fast8_t s_ail_note_halftone[] = {
|
|
|
|
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
|
|
|
|
0x08, 0x09, 0x0a, 0x0b, 0x00, 0x01, 0x02, 0x03,
|
|
|
|
0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b,
|
|
|
|
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
|
|
|
|
0x08, 0x09, 0x0a, 0x0b, 0x00, 0x01, 0x02, 0x03,
|
|
|
|
0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b,
|
|
|
|
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
|
|
|
|
0x08, 0x09, 0x0a, 0x0b, 0x00, 0x01, 0x02, 0x03,
|
|
|
|
0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b,
|
|
|
|
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
|
|
|
|
0x08, 0x09, 0x0a, 0x0b, 0x00, 0x01, 0x02, 0x03,
|
|
|
|
0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b
|
|
|
|
};
|
|
|
|
|
|
|
|
static inline double s_ailFreq(double tone)
|
|
|
|
{
|
|
|
|
int_fast32_t note = (int_fast32_t)(tone);
|
|
|
|
double bendDec = tone - (int)tone; // 0.0 ± 1.0 - one halftone
|
|
|
|
int_fast32_t pitch;
|
|
|
|
uint_fast16_t freq;
|
|
|
|
int_fast32_t octave;
|
|
|
|
int_fast32_t octaveOffset = 0;
|
|
|
|
uint_fast8_t halftones;
|
|
|
|
|
|
|
|
pitch = (int_fast32_t)(bendDec * 4096) + 8192; // convert to MIDI standard value
|
|
|
|
pitch = ((pitch - 0x2000) / 0x20) * 2;
|
|
|
|
|
|
|
|
note -= 12;
|
|
|
|
|
|
|
|
while(note < 0)
|
|
|
|
{
|
|
|
|
octaveOffset--;
|
|
|
|
note += 12;
|
|
|
|
}
|
|
|
|
while(note > 95)
|
|
|
|
{
|
|
|
|
octaveOffset++;
|
|
|
|
note -= 12;
|
|
|
|
}
|
|
|
|
|
|
|
|
pitch += (((uint_fast8_t)note) << 8) + 8;
|
|
|
|
pitch /= 16;
|
|
|
|
while (pitch < 12 * 16) {
|
|
|
|
pitch += 12 * 16;
|
|
|
|
}
|
|
|
|
while (pitch > 96 * 16 - 1) {
|
|
|
|
pitch -= 12 * 16;
|
|
|
|
}
|
|
|
|
|
|
|
|
halftones = (s_ail_note_halftone[pitch >> 4] << 4) + (pitch & 0x0f);
|
|
|
|
freq = s_ail_freqtable[halftones];
|
|
|
|
octave = s_ail_note_octave[pitch >> 4];
|
|
|
|
|
|
|
|
if((freq & 0x8000) == 0)
|
|
|
|
{
|
|
|
|
if (octave > 0) {
|
|
|
|
octave--;
|
|
|
|
} else {
|
|
|
|
freq /= 2;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
freq &= 0x3FF;
|
|
|
|
|
|
|
|
octave += octaveOffset;
|
|
|
|
|
|
|
|
while(octave > 0)
|
|
|
|
{
|
|
|
|
freq *= 2;
|
|
|
|
octave -= 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return freq;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
enum
|
|
|
|
{
|
|
|
|
MasterVolumeDefault = 127
|
|
|
|
};
|
|
|
|
|
|
|
|
enum
|
|
|
|
{
|
|
|
|
OPL_PANNING_LEFT = 0x10,
|
|
|
|
OPL_PANNING_RIGHT = 0x20,
|
|
|
|
OPL_PANNING_BOTH = 0x30
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
static OplInstMeta makeEmptyInstrument()
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
OplInstMeta ins;
|
|
|
|
memset(&ins, 0, sizeof(OplInstMeta));
|
|
|
|
ins.flags = OplInstMeta::Flag_NoSound;
|
2020-01-02 16:52:30 +00:00
|
|
|
return ins;
|
|
|
|
}
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
const OplInstMeta OPL3::m_emptyInstrument = makeEmptyInstrument();
|
2020-01-02 16:52:30 +00:00
|
|
|
|
|
|
|
OPL3::OPL3() :
|
|
|
|
m_numChips(1),
|
|
|
|
m_numFourOps(0),
|
|
|
|
m_deepTremoloMode(false),
|
|
|
|
m_deepVibratoMode(false),
|
|
|
|
m_rhythmMode(false),
|
|
|
|
m_softPanning(false),
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
m_masterVolume(MasterVolumeDefault),
|
2020-01-02 16:52:30 +00:00
|
|
|
m_musicMode(MODE_MIDI),
|
|
|
|
m_volumeScale(VOLUME_Generic)
|
|
|
|
{
|
|
|
|
m_insBankSetup.volumeModel = OPL3::VOLUME_Generic;
|
|
|
|
m_insBankSetup.deepTremolo = false;
|
|
|
|
m_insBankSetup.deepVibrato = false;
|
|
|
|
m_insBankSetup.scaleModulators = false;
|
|
|
|
|
|
|
|
#ifdef DISABLE_EMBEDDED_BANKS
|
|
|
|
m_embeddedBank = CustomBankTag;
|
|
|
|
#else
|
|
|
|
setEmbeddedBank(0);
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
OPL3::~OPL3()
|
|
|
|
{
|
|
|
|
#ifdef ADLMIDI_HW_OPL
|
|
|
|
silenceAll();
|
|
|
|
writeRegI(0, 0x0BD, 0);
|
|
|
|
writeRegI(0, 0x104, 0);
|
|
|
|
writeRegI(0, 0x105, 0);
|
|
|
|
silenceAll();
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2020-01-02 16:52:30 +00:00
|
|
|
bool OPL3::setupLocked()
|
|
|
|
{
|
|
|
|
return (m_musicMode == MODE_CMF ||
|
|
|
|
m_musicMode == MODE_IMF ||
|
|
|
|
m_musicMode == MODE_RSXX);
|
|
|
|
}
|
|
|
|
|
|
|
|
void OPL3::setEmbeddedBank(uint32_t bank)
|
|
|
|
{
|
|
|
|
#ifndef DISABLE_EMBEDDED_BANKS
|
|
|
|
m_embeddedBank = bank;
|
|
|
|
//Embedded banks are supports 128:128 GM set only
|
|
|
|
m_insBanks.clear();
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(bank >= static_cast<uint32_t>(g_embeddedBanksCount))
|
2020-01-02 16:52:30 +00:00
|
|
|
return;
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
const BanksDump::BankEntry &bankEntry = g_embeddedBanks[m_embeddedBank];
|
|
|
|
m_insBankSetup.deepTremolo = ((bankEntry.bankSetup >> 8) & 0x01) != 0;
|
|
|
|
m_insBankSetup.deepVibrato = ((bankEntry.bankSetup >> 8) & 0x02) != 0;
|
|
|
|
m_insBankSetup.volumeModel = (bankEntry.bankSetup & 0xFF);
|
|
|
|
m_insBankSetup.scaleModulators = false;
|
2020-01-02 16:52:30 +00:00
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
for(int ss = 0; ss < 2; ss++)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
bank_count_t maxBanks = ss ? bankEntry.banksPercussionCount : bankEntry.banksMelodicCount;
|
|
|
|
bank_count_t banksOffset = ss ? bankEntry.banksOffsetPercussive : bankEntry.banksOffsetMelodic;
|
|
|
|
|
|
|
|
for(bank_count_t bankID = 0; bankID < maxBanks; bankID++)
|
|
|
|
{
|
|
|
|
size_t bankIndex = g_embeddedBanksMidiIndex[banksOffset + bankID];
|
|
|
|
const BanksDump::MidiBank &bankData = g_embeddedBanksMidi[bankIndex];
|
|
|
|
size_t bankMidiIndex = static_cast<size_t>((bankData.msb * 256) + bankData.lsb) + (ss ? static_cast<size_t>(PercussionTag) : 0);
|
|
|
|
Bank &bankTarget = m_insBanks[bankMidiIndex];
|
|
|
|
|
|
|
|
for(size_t instId = 0; instId < 128; instId++)
|
|
|
|
{
|
|
|
|
midi_bank_idx_t instIndex = bankData.insts[instId];
|
|
|
|
if(instIndex < 0)
|
|
|
|
{
|
|
|
|
bankTarget.ins[instId].flags = OplInstMeta::Flag_NoSound;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
BanksDump::InstrumentEntry instIn = g_embeddedBanksInstruments[instIndex];
|
|
|
|
OplInstMeta &instOut = bankTarget.ins[instId];
|
|
|
|
|
|
|
|
adlFromInstrument(instIn, instOut);
|
|
|
|
}
|
|
|
|
}
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
|
2020-01-02 16:52:30 +00:00
|
|
|
#else
|
|
|
|
ADL_UNUSED(bank);
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
|
|
|
void OPL3::writeReg(size_t chip, uint16_t address, uint8_t value)
|
|
|
|
{
|
|
|
|
#ifdef ADLMIDI_HW_OPL
|
|
|
|
ADL_UNUSED(chip);
|
|
|
|
unsigned o = address >> 8;
|
|
|
|
unsigned port = OPLBase + o * 2;
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
# ifdef __DJGPP__
|
2020-01-02 16:52:30 +00:00
|
|
|
outportb(port, address);
|
|
|
|
for(unsigned c = 0; c < 6; ++c) inportb(port);
|
|
|
|
outportb(port + 1, value);
|
|
|
|
for(unsigned c = 0; c < 35; ++c) inportb(port);
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
# endif
|
2020-01-02 16:52:30 +00:00
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
# ifdef __WATCOMC__
|
2020-01-02 16:52:30 +00:00
|
|
|
outp(port, address);
|
|
|
|
for(uint16_t c = 0; c < 6; ++c) inp(port);
|
|
|
|
outp(port + 1, value);
|
|
|
|
for(uint16_t c = 0; c < 35; ++c) inp(port);
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
# endif//__WATCOMC__
|
2020-01-02 16:52:30 +00:00
|
|
|
|
|
|
|
#else//ADLMIDI_HW_OPL
|
|
|
|
m_chips[chip]->writeReg(address, value);
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
|
|
|
void OPL3::writeRegI(size_t chip, uint32_t address, uint32_t value)
|
|
|
|
{
|
|
|
|
#ifdef ADLMIDI_HW_OPL
|
|
|
|
writeReg(chip, static_cast<uint16_t>(address), static_cast<uint8_t>(value));
|
|
|
|
#else//ADLMIDI_HW_OPL
|
|
|
|
m_chips[chip]->writeReg(static_cast<uint16_t>(address), static_cast<uint8_t>(value));
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
|
|
|
void OPL3::writePan(size_t chip, uint32_t address, uint32_t value)
|
|
|
|
{
|
|
|
|
#ifndef ADLMIDI_HW_OPL
|
|
|
|
m_chips[chip]->writePan(static_cast<uint16_t>(address), static_cast<uint8_t>(value));
|
|
|
|
#else
|
|
|
|
ADL_UNUSED(chip);
|
|
|
|
ADL_UNUSED(address);
|
|
|
|
ADL_UNUSED(value);
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void OPL3::noteOff(size_t c)
|
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
size_t chip = c / NUM_OF_CHANNELS, cc = c % NUM_OF_CHANNELS;
|
2020-01-02 16:52:30 +00:00
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(cc >= OPL3_CHANNELS_RHYTHM_BASE)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
m_regBD[chip] &= ~(0x10 >> (cc - OPL3_CHANNELS_RHYTHM_BASE));
|
2020-01-02 16:52:30 +00:00
|
|
|
writeRegI(chip, 0xBD, m_regBD[chip]);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
writeRegI(chip, 0xB0 + g_channelsMap[cc], m_keyBlockFNumCache[c] & 0xDF);
|
|
|
|
}
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
void OPL3::noteOn(size_t c1, size_t c2, double tone)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
size_t chip = c1 / NUM_OF_CHANNELS, cc1 = c1 % NUM_OF_CHANNELS, cc2 = c2 % NUM_OF_CHANNELS;
|
2020-01-02 16:52:30 +00:00
|
|
|
uint32_t octave = 0, ftone = 0, mul_offset = 0;
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
// Hertz range: 0..131071
|
|
|
|
double hertz;
|
|
|
|
|
|
|
|
// Use different frequency formulas in depend on a volume model
|
|
|
|
switch(m_volumeScale)
|
|
|
|
{
|
|
|
|
case VOLUME_DMX:
|
|
|
|
case VOLUME_DMX_FIXED:
|
|
|
|
hertz = s_dmxFreq(tone);
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_APOGEE:
|
|
|
|
case VOLUME_APOGEE_FIXED:
|
|
|
|
hertz = s_apogeeFreq(tone);
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_9X:
|
|
|
|
case VOLUME_9X_GENERIC_FM:
|
|
|
|
hertz = s_9xFreq(tone);
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_HMI:
|
|
|
|
case VOLUME_HMI_OLD:
|
|
|
|
hertz = s_hmiFreq(tone);
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_AIL:
|
|
|
|
hertz = s_ailFreq(tone);
|
|
|
|
break;
|
|
|
|
|
|
|
|
default:
|
|
|
|
hertz = s_commonFreq(tone);
|
|
|
|
}
|
2020-01-02 16:52:30 +00:00
|
|
|
|
|
|
|
if(hertz < 0)
|
|
|
|
return;
|
|
|
|
|
|
|
|
//Basic range until max of octaves reaching
|
|
|
|
while((hertz >= 1023.5) && (octave < 0x1C00))
|
|
|
|
{
|
|
|
|
hertz /= 2.0; // Calculate octave
|
|
|
|
octave += 0x400;
|
|
|
|
}
|
|
|
|
//Extended range, rely on frequency multiplication increment
|
|
|
|
while(hertz >= 1022.75)
|
|
|
|
{
|
|
|
|
hertz /= 2.0; // Calculate octave
|
|
|
|
mul_offset++;
|
|
|
|
}
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
ftone = octave + static_cast<uint32_t>(hertz /*+ 0.5*/);
|
2020-01-02 16:52:30 +00:00
|
|
|
uint32_t chn = g_channelsMap[cc1];
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
const OplTimbre &patch1 = m_insCache[c1];
|
|
|
|
const OplTimbre &patch2 = m_insCache[c2 < m_insCache.size() ? c2 : 0];
|
2020-01-02 16:52:30 +00:00
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(cc1 < OPL3_CHANNELS_RHYTHM_BASE)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
|
|
|
ftone += 0x2000u; /* Key-ON [KON] */
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
const bool natural_4op = (m_channelCategory[c1] == ChanCat_4op_First);
|
2020-01-02 16:52:30 +00:00
|
|
|
const size_t opsCount = natural_4op ? 4 : 2;
|
|
|
|
const uint16_t op_addr[4] =
|
|
|
|
{
|
|
|
|
g_operatorsMap[cc1 * 2 + 0], g_operatorsMap[cc1 * 2 + 1],
|
|
|
|
g_operatorsMap[cc2 * 2 + 0], g_operatorsMap[cc2 * 2 + 1]
|
|
|
|
};
|
|
|
|
const uint32_t ops[4] =
|
|
|
|
{
|
|
|
|
patch1.modulator_E862 & 0xFF,
|
|
|
|
patch1.carrier_E862 & 0xFF,
|
|
|
|
patch2.modulator_E862 & 0xFF,
|
|
|
|
patch2.carrier_E862 & 0xFF
|
|
|
|
};
|
|
|
|
|
|
|
|
for(size_t op = 0; op < opsCount; op++)
|
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(op_addr[op] == 0xFFF)
|
|
|
|
continue;
|
2020-01-02 16:52:30 +00:00
|
|
|
if(mul_offset > 0)
|
|
|
|
{
|
|
|
|
uint32_t dt = ops[op] & 0xF0;
|
|
|
|
uint32_t mul = ops[op] & 0x0F;
|
|
|
|
if((mul + mul_offset) > 0x0F)
|
|
|
|
{
|
|
|
|
mul_offset = 0;
|
|
|
|
mul = 0x0F;
|
|
|
|
}
|
|
|
|
writeRegI(chip, 0x20 + op_addr[op], (dt | (mul + mul_offset)) & 0xFF);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
writeRegI(chip, 0x20 + op_addr[op], ops[op] & 0xFF);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if(chn != 0xFFF)
|
|
|
|
{
|
|
|
|
writeRegI(chip , 0xA0 + chn, (ftone & 0xFF));
|
|
|
|
writeRegI(chip , 0xB0 + chn, (ftone >> 8));
|
|
|
|
m_keyBlockFNumCache[c1] = (ftone >> 8);
|
|
|
|
}
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(cc1 >= OPL3_CHANNELS_RHYTHM_BASE)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
m_regBD[chip ] |= (0x10 >> (cc1 - OPL3_CHANNELS_RHYTHM_BASE));
|
2020-01-02 16:52:30 +00:00
|
|
|
writeRegI(chip , 0x0BD, m_regBD[chip ]);
|
|
|
|
//x |= 0x800; // for test
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
static inline uint_fast32_t brightnessToOPL(uint_fast32_t brightness)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
double b = static_cast<double>(brightness);
|
|
|
|
double ret = ::round(127.0 * ::sqrt(b * (1.0 / 127.0))) / 2.0;
|
|
|
|
return static_cast<uint_fast32_t>(ret);
|
|
|
|
}
|
|
|
|
|
|
|
|
void OPL3::touchNote(size_t c,
|
|
|
|
uint_fast32_t velocity,
|
|
|
|
uint_fast32_t channelVolume,
|
|
|
|
uint_fast32_t channelExpression,
|
|
|
|
uint_fast32_t brightness,
|
|
|
|
bool isDrum)
|
|
|
|
{
|
|
|
|
size_t chip = c / NUM_OF_CHANNELS, cc = c % NUM_OF_CHANNELS;
|
|
|
|
const OplTimbre &adli = m_insCache[c];
|
|
|
|
size_t cmf_offset = ((m_musicMode == MODE_CMF) && cc >= OPL3_CHANNELS_RHYTHM_BASE) ? 10 : 0;
|
|
|
|
uint16_t o1 = g_operatorsMap[cc * 2 + 0 + cmf_offset];
|
|
|
|
uint16_t o2 = g_operatorsMap[cc * 2 + 1 + cmf_offset];
|
|
|
|
uint8_t srcMod = adli.modulator_40,
|
|
|
|
srcCar = adli.carrier_40;
|
|
|
|
uint32_t mode = 1; // 2-op AM
|
|
|
|
|
|
|
|
uint_fast32_t kslMod = srcMod & 0xC0;
|
|
|
|
uint_fast32_t kslCar = srcCar & 0xC0;
|
|
|
|
uint_fast32_t tlMod = srcMod & 0x3F;
|
|
|
|
uint_fast32_t tlCar = srcCar & 0x3F;
|
|
|
|
|
|
|
|
uint_fast32_t modulator;
|
|
|
|
uint_fast32_t carrier;
|
|
|
|
|
|
|
|
uint_fast32_t volume = 0;
|
|
|
|
uint_fast32_t midiVolume = 0;
|
|
|
|
|
|
|
|
bool do_modulator = false;
|
|
|
|
bool do_carrier = true;
|
|
|
|
|
|
|
|
static const bool do_ops[10][2] =
|
|
|
|
{
|
|
|
|
{ false, true }, /* 2 op FM */
|
|
|
|
{ true, true }, /* 2 op AM */
|
|
|
|
{ false, false }, /* 4 op FM-FM ops 1&2 */
|
|
|
|
{ true, false }, /* 4 op AM-FM ops 1&2 */
|
|
|
|
{ false, true }, /* 4 op FM-AM ops 1&2 */
|
|
|
|
{ true, false }, /* 4 op AM-AM ops 1&2 */
|
|
|
|
{ false, true }, /* 4 op FM-FM ops 3&4 */
|
|
|
|
{ false, true }, /* 4 op AM-FM ops 3&4 */
|
|
|
|
{ false, true }, /* 4 op FM-AM ops 3&4 */
|
|
|
|
{ true, true } /* 4 op AM-AM ops 3&4 */
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
// ------ Mix volumes and compute average ------
|
|
|
|
|
|
|
|
switch(m_volumeScale)
|
|
|
|
{
|
|
|
|
default:
|
|
|
|
case VOLUME_Generic:
|
|
|
|
{
|
|
|
|
volume = velocity * m_masterVolume *
|
|
|
|
channelVolume * channelExpression;
|
|
|
|
|
|
|
|
/* If the channel has arpeggio, the effective volume of
|
|
|
|
* *this* instrument is actually lower due to timesharing.
|
|
|
|
* To compensate, add extra volume that corresponds to the
|
|
|
|
* time this note is *not* heard.
|
|
|
|
* Empirical tests however show that a full equal-proportion
|
|
|
|
* increment sounds wrong. Therefore, using the square root.
|
|
|
|
*/
|
|
|
|
//volume = (int)(volume * std::sqrt( (double) ch[c].users.size() ));
|
|
|
|
const double c1 = 11.541560327111707;
|
|
|
|
const double c2 = 1.601379199767093e+02;
|
|
|
|
const uint_fast32_t minVolume = 1108075; // 8725 * 127
|
|
|
|
|
|
|
|
// The formula below: SOLVE(V=127^4 * 2^( (A-63.49999) / 8), A)
|
|
|
|
if(volume > minVolume)
|
|
|
|
{
|
|
|
|
double lv = std::log(static_cast<double>(volume));
|
|
|
|
volume = static_cast<uint_fast32_t>(lv * c1 - c2);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
volume = 0;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_NATIVE:
|
|
|
|
{
|
|
|
|
volume = velocity * channelVolume * channelExpression;
|
|
|
|
// 4096766 = (127 * 127 * 127) / 2
|
|
|
|
volume = (volume * m_masterVolume) / 4096766;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_DMX:
|
|
|
|
case VOLUME_DMX_FIXED:
|
|
|
|
{
|
|
|
|
volume = (channelVolume * channelExpression * m_masterVolume) / 16129;
|
|
|
|
volume = (s_dmx_volume_model[volume] + 1) << 1;
|
|
|
|
volume = (s_dmx_volume_model[(velocity < 128) ? velocity : 127] * volume) >> 9;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_APOGEE:
|
|
|
|
case VOLUME_APOGEE_FIXED:
|
|
|
|
{
|
|
|
|
midiVolume = (channelVolume * channelExpression * m_masterVolume / 16129);
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_9X:
|
|
|
|
{
|
|
|
|
volume = (channelVolume * channelExpression * m_masterVolume) / 16129;
|
|
|
|
volume = s_w9x_sb16_volume_model[volume >> 2];
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_9X_GENERIC_FM:
|
|
|
|
{
|
|
|
|
volume = (channelVolume * channelExpression * m_masterVolume) / 16129;
|
|
|
|
volume = s_w9x_generic_fm_volume_model[volume >> 2];
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_AIL:
|
|
|
|
{
|
|
|
|
midiVolume = (channelVolume * channelExpression) * 2;
|
|
|
|
midiVolume >>= 8;
|
|
|
|
if(midiVolume != 0)
|
|
|
|
midiVolume++;
|
|
|
|
|
|
|
|
velocity = (velocity & 0x7F) >> 3;
|
|
|
|
velocity = s_ail_vel_graph[velocity];
|
|
|
|
|
|
|
|
midiVolume = (midiVolume * velocity) * 2;
|
|
|
|
midiVolume >>= 8;
|
|
|
|
if(midiVolume != 0)
|
|
|
|
midiVolume++;
|
|
|
|
|
|
|
|
if(m_masterVolume < 127)
|
|
|
|
midiVolume = (midiVolume * m_masterVolume) / 127;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case VOLUME_HMI:
|
|
|
|
case VOLUME_HMI_OLD:
|
|
|
|
{
|
|
|
|
volume = (channelVolume * channelExpression * m_masterVolume) / 16129;
|
|
|
|
volume = (((volume * 128) / 127) * velocity) >> 7;
|
|
|
|
volume = s_hmi_volume_table[volume >> 1];
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2020-01-02 16:52:30 +00:00
|
|
|
if(volume > 63)
|
|
|
|
volume = 63;
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(midiVolume > 127)
|
|
|
|
midiVolume = 127;
|
|
|
|
|
2020-01-02 16:52:30 +00:00
|
|
|
|
|
|
|
if(m_channelCategory[c] == ChanCat_Regular ||
|
|
|
|
m_channelCategory[c] == ChanCat_Rhythm_Bass)
|
|
|
|
{
|
|
|
|
mode = adli.feedconn & 1; // 2-op FM or 2-op AM
|
|
|
|
}
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
else if(m_channelCategory[c] == ChanCat_4op_First ||
|
|
|
|
m_channelCategory[c] == ChanCat_4op_Second)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
const OplTimbre *i0, *i1;
|
2020-01-02 16:52:30 +00:00
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(m_channelCategory[c] == ChanCat_4op_First)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
|
|
|
i0 = &adli;
|
|
|
|
i1 = &m_insCache[c + 3];
|
|
|
|
mode = 2; // 4-op xx-xx ops 1&2
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
i0 = &m_insCache[c - 3];
|
|
|
|
i1 = &adli;
|
|
|
|
mode = 6; // 4-op xx-xx ops 3&4
|
|
|
|
}
|
|
|
|
|
|
|
|
mode += (i0->feedconn & 1) + (i1->feedconn & 1) * 2;
|
|
|
|
}
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
do_modulator = do_ops[mode][0] || m_scaleModulators;
|
|
|
|
do_carrier = do_ops[mode][1] || m_scaleModulators;
|
|
|
|
|
|
|
|
// ------ Compute the total level register output data ------
|
2020-01-02 16:52:30 +00:00
|
|
|
|
|
|
|
if(m_musicMode == MODE_RSXX)
|
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
tlCar -= volume / 2;
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
else if(m_volumeScale == Synth::VOLUME_APOGEE ||
|
|
|
|
m_volumeScale == Synth::VOLUME_APOGEE_FIXED)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
// volume = ((64 * (velocity + 0x80)) * volume) >> 15;
|
2020-01-02 16:52:30 +00:00
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(do_carrier)
|
|
|
|
{
|
|
|
|
tlCar = 63 - tlCar;
|
|
|
|
tlCar *= velocity + 0x80;
|
|
|
|
tlCar = (midiVolume * tlCar) >> 15;
|
|
|
|
tlCar = tlCar ^ 63;
|
|
|
|
}
|
2020-01-02 16:52:30 +00:00
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(do_modulator)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
uint_fast32_t mod = tlCar;
|
|
|
|
|
|
|
|
tlMod = 63 - tlMod;
|
|
|
|
tlMod *= velocity + 0x80;
|
|
|
|
|
|
|
|
if(m_volumeScale == Synth::VOLUME_APOGEE_FIXED || mode > 1)
|
|
|
|
mod = tlMod; // Fix the AM voices bug
|
|
|
|
|
|
|
|
// NOTE: Here is a bug of Apogee Sound System that makes modulator
|
|
|
|
// to not work properly on AM instruments. The fix of this bug, you
|
|
|
|
// need to replace the tlCar with tmMod in this formula.
|
|
|
|
// Don't do the bug on 4-op voices.
|
|
|
|
tlMod = (midiVolume * mod) >> 15;
|
|
|
|
|
|
|
|
tlMod ^= 63;
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
}
|
|
|
|
else if(m_volumeScale == Synth::VOLUME_DMX && mode <= 1)
|
|
|
|
{
|
|
|
|
tlCar = (63 - volume);
|
2020-01-02 16:52:30 +00:00
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(do_modulator)
|
|
|
|
{
|
|
|
|
if(tlMod < tlCar)
|
|
|
|
tlMod = tlCar;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else if(m_volumeScale == Synth::VOLUME_9X)
|
|
|
|
{
|
|
|
|
if(do_carrier)
|
|
|
|
tlCar += volume + s_w9x_sb16_volume_model[velocity >> 2];
|
|
|
|
if(do_modulator)
|
|
|
|
tlMod += volume + s_w9x_sb16_volume_model[velocity >> 2];
|
|
|
|
|
|
|
|
if(tlCar > 0x3F)
|
|
|
|
tlCar = 0x3F;
|
|
|
|
if(tlMod > 0x3F)
|
|
|
|
tlMod = 0x3F;
|
|
|
|
}
|
|
|
|
else if(m_volumeScale == Synth::VOLUME_9X_GENERIC_FM)
|
|
|
|
{
|
|
|
|
if(do_carrier)
|
|
|
|
tlCar += volume + s_w9x_generic_fm_volume_model[velocity >> 2];
|
|
|
|
if(do_modulator)
|
|
|
|
tlMod += volume + s_w9x_generic_fm_volume_model[velocity >> 2];
|
|
|
|
|
|
|
|
if(tlCar > 0x3F)
|
|
|
|
tlCar = 0x3F;
|
|
|
|
if(tlMod > 0x3F)
|
|
|
|
tlMod = 0x3F;
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
else if(m_volumeScale == Synth::VOLUME_AIL)
|
|
|
|
{
|
|
|
|
uint_fast32_t v0_val = (~srcMod) & 0x3f;
|
|
|
|
uint_fast32_t v1_val = (~srcCar) & 0x3f;
|
|
|
|
|
|
|
|
if(do_modulator)
|
|
|
|
v0_val = (v0_val * midiVolume) / 127;
|
|
|
|
if(do_carrier)
|
|
|
|
v1_val = (v1_val * midiVolume) / 127;
|
|
|
|
|
|
|
|
tlMod = (~v0_val) & 0x3F;
|
|
|
|
tlCar = (~v1_val) & 0x3F;
|
|
|
|
}
|
|
|
|
else if(m_volumeScale == Synth::VOLUME_HMI)
|
|
|
|
{
|
|
|
|
uint_fast32_t vol;
|
|
|
|
|
|
|
|
if(do_modulator)
|
|
|
|
{
|
|
|
|
vol = (64 - volume) << 1;
|
|
|
|
vol *= (64 - tlMod);
|
|
|
|
tlMod = (8192 - vol) >> 7;
|
|
|
|
}
|
|
|
|
|
|
|
|
if(do_carrier)
|
|
|
|
{
|
|
|
|
vol = (64 - volume) << 1;
|
|
|
|
vol *= (64 - tlCar);
|
|
|
|
tlCar = (8192 - vol) >> 7;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else if(m_volumeScale == Synth::VOLUME_HMI_OLD)
|
|
|
|
{
|
|
|
|
uint_fast32_t vol;
|
|
|
|
|
|
|
|
if(adli.feedconn == 0 && !isDrum)
|
|
|
|
{
|
|
|
|
vol = (channelVolume * channelExpression * 64) / 16129;
|
|
|
|
vol = (((vol * 128) / 127) * velocity) >> 7;
|
|
|
|
vol = s_hmi_volume_table[vol >> 1];
|
|
|
|
|
|
|
|
vol = (64 - vol) << 1;
|
|
|
|
vol *= (64 - tlCar);
|
|
|
|
tlMod = (8192 - vol) >> 7;
|
|
|
|
}
|
|
|
|
|
|
|
|
if(isDrum) // TODO: VERIFY A CORRECTNESS OF THIS!!!
|
|
|
|
vol = s_hmi_volume_table[velocity >> 1];
|
|
|
|
|
|
|
|
vol = (64 - volume) << 1;
|
|
|
|
vol *= (64 - tlCar);
|
|
|
|
tlCar = (8192 - vol) >> 7;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
if(do_modulator)
|
|
|
|
tlMod = 63 - volume + (volume * tlMod) / 63;
|
|
|
|
if(do_carrier)
|
|
|
|
tlCar = 63 - volume + (volume * tlCar) / 63;
|
|
|
|
}
|
|
|
|
|
|
|
|
if(brightness != 127 && !isDrum)
|
|
|
|
{
|
|
|
|
brightness = brightnessToOPL(brightness);
|
|
|
|
if(!do_modulator)
|
|
|
|
tlMod = 63 - brightness + (brightness * tlMod) / 63;
|
|
|
|
if(!do_carrier)
|
|
|
|
tlCar = 63 - brightness + (brightness * tlCar) / 63;
|
|
|
|
}
|
|
|
|
|
|
|
|
modulator = (kslMod & 0xC0) | (tlMod & 63);
|
|
|
|
carrier = (kslCar & 0xC0) | (tlCar & 63);
|
|
|
|
|
|
|
|
if(o1 != 0xFFF)
|
|
|
|
writeRegI(chip, 0x40 + o1, modulator);
|
|
|
|
if(o2 != 0xFFF)
|
|
|
|
writeRegI(chip, 0x40 + o2, carrier);
|
2020-01-02 16:52:30 +00:00
|
|
|
|
|
|
|
// Correct formula (ST3, AdPlug):
|
|
|
|
// 63-((63-(instrvol))/63)*chanvol
|
|
|
|
// Reduces to (tested identical):
|
|
|
|
// 63 - chanvol + chanvol*instrvol/63
|
|
|
|
// Also (slower, floats):
|
|
|
|
// 63 + chanvol * (instrvol / 63.0 - 1)
|
|
|
|
}
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
void OPL3::setPatch(size_t c, const OplTimbre &instrument)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
size_t chip = c / NUM_OF_CHANNELS, cc = c % NUM_OF_CHANNELS;
|
2020-01-02 16:52:30 +00:00
|
|
|
static const uint8_t data[4] = {0x20, 0x60, 0x80, 0xE0};
|
|
|
|
m_insCache[c] = instrument;
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
size_t cmf_offset = ((m_musicMode == MODE_CMF) && (cc >= OPL3_CHANNELS_RHYTHM_BASE)) ? 10 : 0;
|
|
|
|
uint16_t o1 = g_operatorsMap[cc * 2 + 0 + cmf_offset];
|
|
|
|
uint16_t o2 = g_operatorsMap[cc * 2 + 1 + cmf_offset];
|
2020-01-02 16:52:30 +00:00
|
|
|
unsigned x = instrument.modulator_E862, y = instrument.carrier_E862;
|
|
|
|
|
|
|
|
for(size_t a = 0; a < 4; ++a, x >>= 8, y >>= 8)
|
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
if(o1 != 0xFFF)
|
|
|
|
writeRegI(chip, data[a] + o1, x & 0xFF);
|
2020-01-02 16:52:30 +00:00
|
|
|
if(o2 != 0xFFF)
|
|
|
|
writeRegI(chip, data[a] + o2, y & 0xFF);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void OPL3::setPan(size_t c, uint8_t value)
|
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
size_t chip = c / NUM_OF_CHANNELS, cc = c % NUM_OF_CHANNELS;
|
|
|
|
if(g_channelsMapPan[cc] != 0xFFF)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
|
|
|
#ifndef ADLMIDI_HW_OPL
|
|
|
|
if (m_softPanning)
|
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
writePan(chip, g_channelsMapPan[cc], value);
|
|
|
|
writeRegI(chip, 0xC0 + g_channelsMapPan[cc], m_insCache[c].feedconn | OPL_PANNING_BOTH);
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
#endif
|
|
|
|
int panning = 0;
|
|
|
|
if(value < 64 + 32) panning |= OPL_PANNING_LEFT;
|
|
|
|
if(value >= 64 - 32) panning |= OPL_PANNING_RIGHT;
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
writePan(chip, g_channelsMapPan[cc], 64);
|
|
|
|
writeRegI(chip, 0xC0 + g_channelsMapPan[cc], m_insCache[c].feedconn | panning);
|
2020-01-02 16:52:30 +00:00
|
|
|
#ifndef ADLMIDI_HW_OPL
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void OPL3::silenceAll() // Silence all OPL channels.
|
|
|
|
{
|
|
|
|
for(size_t c = 0; c < m_numChannels; ++c)
|
|
|
|
{
|
|
|
|
noteOff(c);
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
touchNote(c, 0, 0, 0);
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void OPL3::updateChannelCategories()
|
|
|
|
{
|
|
|
|
const uint32_t fours = m_numFourOps;
|
|
|
|
|
|
|
|
for(uint32_t chip = 0, fours_left = fours; chip < m_numChips; ++chip)
|
|
|
|
{
|
|
|
|
m_regBD[chip] = (m_deepTremoloMode * 0x80 + m_deepVibratoMode * 0x40 + m_rhythmMode * 0x20);
|
|
|
|
writeRegI(chip, 0x0BD, m_regBD[chip]);
|
|
|
|
uint32_t fours_this_chip = std::min(fours_left, static_cast<uint32_t>(6u));
|
|
|
|
writeRegI(chip, 0x104, (1 << fours_this_chip) - 1);
|
|
|
|
fours_left -= fours_this_chip;
|
|
|
|
}
|
|
|
|
|
|
|
|
if(!m_rhythmMode)
|
|
|
|
{
|
|
|
|
for(size_t a = 0, n = m_numChips; a < n; ++a)
|
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
for(size_t b = 0; b < NUM_OF_CHANNELS; ++b)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
m_channelCategory[a * NUM_OF_CHANNELS + b] =
|
|
|
|
(b >= OPL3_CHANNELS_RHYTHM_BASE) ? ChanCat_Rhythm_Secondary : ChanCat_Regular;
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else
|
|
|
|
{
|
|
|
|
for(size_t a = 0, n = m_numChips; a < n; ++a)
|
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
for(size_t b = 0; b < NUM_OF_CHANNELS; ++b)
|
2020-01-02 16:52:30 +00:00
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
m_channelCategory[a * NUM_OF_CHANNELS + b] =
|
|
|
|
(b >= OPL3_CHANNELS_RHYTHM_BASE) ? static_cast<ChanCat>(ChanCat_Rhythm_Bass + (b - OPL3_CHANNELS_RHYTHM_BASE)) :
|
|
|
|
(b >= 6 && b < 9) ? ChanCat_Rhythm_Secondary : ChanCat_Regular;
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
uint32_t nextfour = 0;
|
|
|
|
for(uint32_t a = 0; a < fours; ++a)
|
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
m_channelCategory[nextfour] = ChanCat_4op_First;
|
|
|
|
m_channelCategory[nextfour + 3] = ChanCat_4op_Second;
|
2020-01-02 16:52:30 +00:00
|
|
|
|
|
|
|
switch(a % 6)
|
|
|
|
{
|
|
|
|
case 0:
|
|
|
|
case 1:
|
|
|
|
nextfour += 1;
|
|
|
|
break;
|
|
|
|
case 2:
|
|
|
|
nextfour += 9 - 2;
|
|
|
|
break;
|
|
|
|
case 3:
|
|
|
|
case 4:
|
|
|
|
nextfour += 1;
|
|
|
|
break;
|
|
|
|
case 5:
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
nextfour += NUM_OF_CHANNELS - 9 - 2;
|
2020-01-02 16:52:30 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/**/
|
|
|
|
/*
|
|
|
|
In two-op mode, channels 0..8 go as follows:
|
|
|
|
Op1[port] Op2[port]
|
|
|
|
Channel 0: 00 00 03 03
|
|
|
|
Channel 1: 01 01 04 04
|
|
|
|
Channel 2: 02 02 05 05
|
|
|
|
Channel 3: 06 08 09 0B
|
|
|
|
Channel 4: 07 09 10 0C
|
|
|
|
Channel 5: 08 0A 11 0D
|
|
|
|
Channel 6: 12 10 15 13
|
|
|
|
Channel 7: 13 11 16 14
|
|
|
|
Channel 8: 14 12 17 15
|
|
|
|
In four-op mode, channels 0..8 go as follows:
|
|
|
|
Op1[port] Op2[port] Op3[port] Op4[port]
|
|
|
|
Channel 0: 00 00 03 03 06 08 09 0B
|
|
|
|
Channel 1: 01 01 04 04 07 09 10 0C
|
|
|
|
Channel 2: 02 02 05 05 08 0A 11 0D
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
Channel 3: CHANNEL 0 SECONDARY
|
|
|
|
Channel 4: CHANNEL 1 SECONDARY
|
|
|
|
Channel 5: CHANNEL 2 SECONDARY
|
2020-01-02 16:52:30 +00:00
|
|
|
Channel 6: 12 10 15 13
|
|
|
|
Channel 7: 13 11 16 14
|
|
|
|
Channel 8: 14 12 17 15
|
|
|
|
Same goes principally for channels 9-17 respectively.
|
|
|
|
*/
|
|
|
|
}
|
|
|
|
|
|
|
|
void OPL3::commitDeepFlags()
|
|
|
|
{
|
|
|
|
for(size_t chip = 0; chip < m_numChips; ++chip)
|
|
|
|
{
|
|
|
|
m_regBD[chip] = (m_deepTremoloMode * 0x80 + m_deepVibratoMode * 0x40 + m_rhythmMode * 0x20);
|
|
|
|
writeRegI(chip, 0x0BD, m_regBD[chip]);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void OPL3::setVolumeScaleModel(ADLMIDI_VolumeModels volumeModel)
|
|
|
|
{
|
|
|
|
switch(volumeModel)
|
|
|
|
{
|
|
|
|
case ADLMIDI_VolumeModel_AUTO://Do nothing until restart playing
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_Generic:
|
|
|
|
m_volumeScale = OPL3::VOLUME_Generic;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_NativeOPL3:
|
|
|
|
m_volumeScale = OPL3::VOLUME_NATIVE;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_DMX:
|
|
|
|
m_volumeScale = OPL3::VOLUME_DMX;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_APOGEE:
|
|
|
|
m_volumeScale = OPL3::VOLUME_APOGEE;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_9X:
|
|
|
|
m_volumeScale = OPL3::VOLUME_9X;
|
|
|
|
break;
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_DMX_Fixed:
|
|
|
|
m_volumeScale = OPL3::VOLUME_DMX_FIXED;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_APOGEE_Fixed:
|
|
|
|
m_volumeScale = OPL3::VOLUME_APOGEE_FIXED;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_AIL:
|
|
|
|
m_volumeScale = OPL3::VOLUME_AIL;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_9X_GENERIC_FM:
|
|
|
|
m_volumeScale = OPL3::VOLUME_9X_GENERIC_FM;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_HMI:
|
|
|
|
m_volumeScale = OPL3::VOLUME_HMI;
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ADLMIDI_VolumeModel_HMI_OLD:
|
|
|
|
m_volumeScale = OPL3::VOLUME_HMI_OLD;
|
|
|
|
break;
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
ADLMIDI_VolumeModels OPL3::getVolumeScaleModel()
|
|
|
|
{
|
|
|
|
switch(m_volumeScale)
|
|
|
|
{
|
|
|
|
default:
|
|
|
|
case OPL3::VOLUME_Generic:
|
|
|
|
return ADLMIDI_VolumeModel_Generic;
|
|
|
|
case OPL3::VOLUME_NATIVE:
|
|
|
|
return ADLMIDI_VolumeModel_NativeOPL3;
|
|
|
|
case OPL3::VOLUME_DMX:
|
|
|
|
return ADLMIDI_VolumeModel_DMX;
|
|
|
|
case OPL3::VOLUME_APOGEE:
|
|
|
|
return ADLMIDI_VolumeModel_APOGEE;
|
|
|
|
case OPL3::VOLUME_9X:
|
|
|
|
return ADLMIDI_VolumeModel_9X;
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
case OPL3::VOLUME_DMX_FIXED:
|
|
|
|
return ADLMIDI_VolumeModel_DMX_Fixed;
|
|
|
|
case OPL3::VOLUME_APOGEE_FIXED:
|
|
|
|
return ADLMIDI_VolumeModel_APOGEE_Fixed;
|
|
|
|
case OPL3::VOLUME_AIL:
|
|
|
|
return ADLMIDI_VolumeModel_AIL;
|
|
|
|
case OPL3::VOLUME_9X_GENERIC_FM:
|
|
|
|
return ADLMIDI_VolumeModel_9X_GENERIC_FM;
|
|
|
|
case OPL3::VOLUME_HMI:
|
|
|
|
return ADLMIDI_VolumeModel_HMI;
|
|
|
|
case OPL3::VOLUME_HMI_OLD:
|
|
|
|
return ADLMIDI_VolumeModel_HMI_OLD;
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifndef ADLMIDI_HW_OPL
|
|
|
|
void OPL3::clearChips()
|
|
|
|
{
|
|
|
|
for(size_t i = 0; i < m_chips.size(); i++)
|
|
|
|
m_chips[i].reset(NULL);
|
|
|
|
m_chips.clear();
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
void OPL3::reset(int emulator, unsigned long PCM_RATE, void *audioTickHandler)
|
|
|
|
{
|
|
|
|
#ifndef ADLMIDI_HW_OPL
|
|
|
|
clearChips();
|
|
|
|
#else
|
|
|
|
(void)emulator;
|
|
|
|
(void)PCM_RATE;
|
|
|
|
#endif
|
|
|
|
#if !defined(ADLMIDI_AUDIO_TICK_HANDLER)
|
|
|
|
(void)audioTickHandler;
|
|
|
|
#endif
|
|
|
|
m_insCache.clear();
|
|
|
|
m_keyBlockFNumCache.clear();
|
|
|
|
m_regBD.clear();
|
|
|
|
|
|
|
|
#ifndef ADLMIDI_HW_OPL
|
|
|
|
m_chips.resize(m_numChips, AdlMIDI_SPtr<OPLChipBase>());
|
|
|
|
#endif
|
|
|
|
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
const struct OplTimbre defaultInsCache = { 0x1557403,0x005B381, 0x49,0x80, 0x4, +0 };
|
|
|
|
m_numChannels = m_numChips * NUM_OF_CHANNELS;
|
2020-01-02 16:52:30 +00:00
|
|
|
m_insCache.resize(m_numChannels, defaultInsCache);
|
|
|
|
m_keyBlockFNumCache.resize(m_numChannels, 0);
|
|
|
|
m_regBD.resize(m_numChips, 0);
|
|
|
|
m_channelCategory.resize(m_numChannels, 0);
|
|
|
|
|
|
|
|
for(size_t p = 0, a = 0; a < m_numChips; ++a)
|
|
|
|
{
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
for(size_t b = 0; b < OPL3_CHANNELS_RHYTHM_BASE; ++b)
|
|
|
|
m_channelCategory[p++] = ChanCat_Regular;
|
|
|
|
for(size_t b = 0; b < NUM_OF_RM_CHANNELS; ++b)
|
|
|
|
m_channelCategory[p++] = ChanCat_Rhythm_Secondary;
|
2020-01-02 16:52:30 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static const uint16_t data[] =
|
|
|
|
{
|
|
|
|
0x004, 96, 0x004, 128, // Pulse timer
|
|
|
|
0x105, 0, 0x105, 1, 0x105, 0, // Pulse OPL3 enable
|
|
|
|
0x001, 32, 0x105, 1 // Enable wave, OPL3 extensions
|
|
|
|
};
|
|
|
|
// size_t fours = m_numFourOps;
|
|
|
|
|
|
|
|
for(size_t i = 0; i < m_numChips; ++i)
|
|
|
|
{
|
|
|
|
#ifndef ADLMIDI_HW_OPL
|
|
|
|
OPLChipBase *chip;
|
|
|
|
switch(emulator)
|
|
|
|
{
|
|
|
|
default:
|
|
|
|
assert(false);
|
|
|
|
abort();
|
|
|
|
#ifndef ADLMIDI_DISABLE_NUKED_EMULATOR
|
|
|
|
case ADLMIDI_EMU_NUKED: /* Latest Nuked OPL3 */
|
|
|
|
chip = new NukedOPL3;
|
|
|
|
break;
|
|
|
|
case ADLMIDI_EMU_NUKED_174: /* Old Nuked OPL3 1.4.7 modified and optimized */
|
|
|
|
chip = new NukedOPL3v174;
|
|
|
|
break;
|
|
|
|
#endif
|
|
|
|
#ifndef ADLMIDI_DISABLE_DOSBOX_EMULATOR
|
|
|
|
case ADLMIDI_EMU_DOSBOX:
|
|
|
|
chip = new DosBoxOPL3;
|
|
|
|
break;
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
#endif
|
|
|
|
#ifndef ADLMIDI_DISABLE_OPAL_EMULATOR
|
|
|
|
case ADLMIDI_EMU_OPAL:
|
|
|
|
chip = new OpalOPL3;
|
|
|
|
break;
|
|
|
|
#endif
|
|
|
|
#ifndef ADLMIDI_DISABLE_JAVA_EMULATOR
|
|
|
|
case ADLMIDI_EMU_JAVA:
|
|
|
|
chip = new JavaOPL3;
|
|
|
|
break;
|
2020-01-02 16:52:30 +00:00
|
|
|
#endif
|
|
|
|
}
|
|
|
|
m_chips[i].reset(chip);
|
|
|
|
chip->setChipId((uint32_t)i);
|
|
|
|
chip->setRate((uint32_t)PCM_RATE);
|
|
|
|
if(m_runAtPcmRate)
|
|
|
|
chip->setRunningAtPcmRate(true);
|
|
|
|
# if defined(ADLMIDI_AUDIO_TICK_HANDLER)
|
|
|
|
chip->setAudioTickHandlerInstance(audioTickHandler);
|
|
|
|
# endif
|
|
|
|
#endif // ADLMIDI_HW_OPL
|
|
|
|
|
|
|
|
/* Clean-up channels from any playing junk sounds */
|
Update libADLMIDI up to 1.5.0
## 1.5.0 2020-09-28
* Drum note length expanding is now supported in real-time mode (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Channels manager has been improved (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Nuked OPL3 1.8 emulator got some optimizations ported from 1.7 where they are was applied previously (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Reworked rhythm-mode percussions system, WOPL banks with rhythm-mode percussions
* Added Public Domain Opal OPL3 emulator made by Reality (a team who originally made the Reality Adlib Tracker) (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Added LGPL licensed JavaOPL3 emulator made by Robson Cozendey in Java and later rewritten into C++ for GZDoom (Thanks to [Jean Pierre Cimalando](https://github.com/jpcima) for a work!)
* Fully rewritten an embedded bank database format, embedded banks now supports a wider set (more than 127:127 instruments in one bank)
* Improved accuracy of the DMX volume model, include the buggy AM interpretation
* Improved accuracy of Apogee volume model, include the bug of AM instruments
* Improved accuracy of the Win9X volume model
* Removed C++ extras. C++-bounded instruments tester is useless since a real-time MIDI API can completely replace it
* Added AIL volume model
* Added Generic FM variant of Win9X volume model
* Fixed an incorrect work of CC-121 (See https://github.com/Wohlstand/libADLMIDI/issues/227 for details)
* Added HMI volume model (Thanks to [Alexey Khokholov](https://github.com/nukeykt) for help with research!)
* Added frequency models, assigned to every volume model: AIL, HMI, DMX, Apogee, 9X, and the Generic formula
2020-09-28 18:17:46 +00:00
|
|
|
for(size_t a = 0; a < OPL3_CHANNELS_RHYTHM_BASE; ++a)
|
2020-01-02 16:52:30 +00:00
|
|
|
writeRegI(i, 0xB0 + g_channelsMap[a], 0x00);
|
|
|
|
for(size_t a = 0; a < sizeof(data) / sizeof(*data); a += 2)
|
|
|
|
writeRegI(i, data[a], (data[a + 1]));
|
|
|
|
}
|
|
|
|
|
|
|
|
updateChannelCategories();
|
|
|
|
silenceAll();
|
|
|
|
}
|