Forked from Alexey Khohkolov's (khokh2001) OPL3 synth. This intends to improve onto the base driver to include extended MIDI support and other improvements to the playback component.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Compilation issues with MinGW toolchain due to strictness of linker scope and typing. MSVC does a lot of hand holding when it comes to commonly used type macros and whatnot, but to allow for code portability this needs to be considered..
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Self-explanatory. To be investigated.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Need to handle some GS ADSR commands via NRPN which are handled via CC for XG.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
This mainly affect 2x2op/4op patches, but enabling mono mode will invalidate notation from second voices. Also should retrigger instead of glissando.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Strange patch mapping behaviour on note-ons after a long period of playback.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Original patch parameters are being restored when Opl3_CalcVolume() is called, presenting inconsistencies in audible playback.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Some voices are being output as if 4-op mode is still on at an intermittent basis due to the toggle not being flagged on time on next allocation. To determine the cause.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Currently will only work in legato by forgetting past notes. This may be unplayable for performance. Consider using a note history stack structure.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Currently using waveOut (inherited from Alexey's source). Need to use implementation as demonstrated by kode54's BASSMIDI driver.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
This will facilitate the following:
All shall use the same interface function for writing to OPL registers.
For some reason when including opl3emu.dll in the Drivers32 registry entry so that it appears as a selectable MIDI output device, from time to time I would encounter issues where random applications would fail to load (eg: vmware.exe) or otherwise crash out of nowhere when using specific options (eg: winamp.exe property pages for other DLL plugins). This is likely due to a resource leak/other low-level WinMM API abuse that was not accounted for.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Need a better way to keep track of allocating notes given patch changes.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Appears to be issues when dealing with mono-legato notes and modulation.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Good opportunity to add something like a pitch falling speed of sorts, to allow for percussive sounds without having to manually map pitch bends.
Prerequisites: requires method of updating/sending commands every 441 (100Hz) samples at the very least, much like LFO modwheel vibrato.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
WinFM OPL3 using Fat Man patches has always been able to play CANYON.MID flawlessly, but recent changes in note allocation logic has now caused it to play back with dropped notes. This is unacceptable output.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Provide a configuration utility which will be the central user interface for modifying the behaviour of the synth driver. (ie: toggle enhancements, bank import, logging, playback modes etc.)
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Provide a method to play back MIDI files using 4-operator patches. To determine:
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Eventually, there will be features implemented that are specific to this driver, which will require documenting their usage.
Original report by Jean Pierre Cimalando (Bitbucket: jpcima, GitHub: jpcima).
The function GenerateResampled of nukedOPL produces an output converted from the internal sample rate of 49716 Hz.
Apparently it's a linear interpolation implemented in fixed point.
Could the sound quality improve in significant manner by providing better resampling methods?
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
A target intention for demonstrating 2x2op patch support is to allow for playback with the Voyetra SuperSAPI! patches.
Current challenges:
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
After refactoring audio component, note offs don't appear to be sent on teardown when playing through hardware. Also with Winamp under XP, it seems timing issues are evident around startup.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Usage on a single-core 1GHz machine under WinXP produces playable but unacceptable performance (ie :80-100% cpu) with hardware playback.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
2x2op patches appears to have problems with transitioning between glissando bends on poly mode, as well as note history bends on mono mode.
Considerations needed on whether or not to use linear semitone-time bend policy as a user preference, as well as glide time range
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Add additional tuning support via Coase/Fine NRPN commands (GM-2/GS/XG compliant)
Coarse (RPN LSB=2, RPN MSB=0) = tones. used as note offset
Fine (RPN LSB=1, RPN MSB=0) = "cents". -/+whole semitone
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Made obvious with the emscripten experiment, large audio buffers made obvious the delay between MIDI events being processed between buffer arrays being rendered. That is, they are being quantized based on the buffer size which is not what we want.
Need to look at options on how to better improve time-based handling of MIDI events.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
CC118 being used to set loop start marker for resultant .VGM log does not appear to work correctly when played back in in_vgm or related.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Provide a method to persist register data to be saved to a .vgm file conforming to at least v1.51 or later specifications. Preferably togglable via configuration to intercept instance in memory, or otherwise by MIDI command similar to S98 logging in VOPMex.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
No effect on patches eg: 81 Square Wave (on Fat Man bank)
Also overflow wrap-around issues on maximum/minimum CC values of CC72/73/74 in some instances.
I use wopl2syx to convert wopl to a header file and then compile it. It's strange to find that the timbre is not 2x2op, but it sounds like 2op, and the other 2op seems to be lost.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Some values are not the same when making changes during a held note, as opposed to changing then issuing a note. In the case of CC11, notes go completely silent.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
ValleyBell's MidiPlay supposedly allows advanced users to use MIDI messages to modify their patches during playback runtime via NRPN messages. To investigate whether this method is suitable, or whether or not this behaviour should be incorporated with SysEx.
Currently not a priority, and certainly unlikely without reorganising the in-memory representation of patches.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Currently it is possible to allocate channels to note on events where output is silent. Implement a way to determine what is considered silent by analysing operator registers, and setting bOn to FALSE where appropriate.)
NB: Working out a method for this will also allow for a method of implementing software-bound modulation wheel vibrato
Original report by Wohlstand (Bitbucket: Wohlstand, GitHub: Wohlstand).
Hello!
I have some bunch of IBK files made by me, and I was used the Jamie O’Connel’s custom FM MIDI Driver to play&record music with it, but it has a poor support of drums (I can't implement full support of them).
That driver (Win 3x / Win 9x):
http://wohlsoft.ru/docs/Sounds/SMBX_OPL/SMBX_OPL_Sounds_src/Software%20for%20FM%20chip/fmsynthJamieOConnel.zip
SMB Timbre tool (DOS 16 bit) which I used to make my mods:
http://wohlsoft.ru/docs/Sounds/SMBX_OPL/SMBX_OPL_Sounds_src/Software%20for%20FM%20chip/sbtimbre.zip
My banks: http://wohlsoft.ru/docs/Sounds/SMBX_OPL/SMBX_OPL_Sounds_src/Software%20for%20FM%20chip/IBK%20banks/
EDIT:
Half year ago I found the ADLMIDI program (which I redesigned into the library) which implements an MIDI-Player over OPL chip emulator and uses a big set of various bank formats. It works light and is not overloads CPU, tested on Pentium IV machine. This also supports overclocking 18-channels 2-op and 6-channels 4-op limits by running multiple copies of chip emulators and then MIDI-player shares notes between all running chip emulators.
Now I got skills and then I quickly made a bank editor for Junglevision format:
https://github.com/Wohlstand/OPL3BankEditor (also on bitbucket too: https://bitbucket.org/Wohlstand/opl3bankeditor)
That format is supported both 2-operator, 4-operator instruments. Then later I have been added more bank formats: DMX, TMB, IBK, BNK (AdLib and HMI), and AIL (also I working on other formats too).
I think, I'll make a new file format which must support:
**EDIT2: ** The new WOPL format has been created
It is already supported by OPL3 Bank Editor and by libADLMIDI:
The full specification is here: https://github.com/Wohlstand/OPL3BankEditor/blob/master/Specifications/WOPL-and-OPLI-Specification.txt
Feel free to use it on your side!
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Currently, for cases such as all 4op voice channels (0,1,2,3,4,5,9,10,11,12,13,14) and 2op voices taking up the rest, new voices appear to only attempt to steal its own last voice like as if it were in mono legato mode. This is undesirable when it should be stealing the oldest note.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Self-explanatory. To be investigated. Refer to bank format included with bisqwit's ADLMIDI.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
It turns out, for banks such as CTMIDI and SAPI!, there appears to be a need for things such as transpose, detune, pan locks and perhaps maybe potentially even retrigger periods. Without altering the patch bank structure (for now), like with the drum bank note/key maps, have something for each patch to assign specific properties.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Key off release presents audible problems with modulation.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Refactor patch retrieval routines to read from an STL map, which should make lookups just as effective but allow for a more programmer-friendly way to expand banks according to the MIDI spec.
eg: Bank Select MSB/LSB categorisation and fallback filters.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Self-explanatory. No special features observed with patch characteristics so far.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
eg: some vgm2mid conversions, Teck's Plok_BreezyBeach.mid, among others.
Need to determine root cause of the crash. Most likely a DIV0 or bad memory access due to race condition.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Utilise GM/GM2/GS/XG reset commands for allocating specific features to be available for MIDI.playback.
eg: GM reset will disable switchable drum banks. Default to GM-2 state.
Original report by James Alan Nguyen (Bitbucket: djtubig-malicex, GitHub: djtubig-malicex).
Allow for handling patches with a second 2-operator pair which is already handled by the existing patch structure. Requires new member fields to keep track of voices that are related.
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.
