DobieStation Git (2020/06/10) is compiled. A young PS2 emulator with plans for an optimized Android port, as well as a fast, accurate, and easy-to-use PC port. A large portion of the PS2's library can boot or get to menus. Some titles can even go in-game, including high-profile ones such as Final Fantasy X and Shadow of the Colossus. Not intended for general use.

DobieStation Git Changelog:
* Merge pull request #406 from Mrlinkwii/master
* Update emotion.cpp
* Merge pull request #333 from Ziemas/audio
* Struct renames
* Delay wav writer file opening until we have data
* Checkbox for WAV audio output
* Dr. Hell's noise generator
* Split out decoding from sample advance loop
* Clarify some things with comments.
* Update key_on/off state at the end of the next T
* Fix some reverb stuff
* Reset interpolation data on key on
* Unify input/output buffer management
* Errors::die on adma bit 2 being set
* Support muting core output
* Don't alter provided reverb offsets
* Add noise to savestate
* Make it actually sound like noise
* Noise generation (idk might change method later)
* Write out the rest of the memout buffers to memory
* Fix more dumb mistakes
* Support reads from ESAH/L
* Sort out some reg confusion
* Fix dumb pitch mod mistake
* Remove leftovers
* Support reads from MMIX
* Write core0 output and read it on core1
* Die on pitch modulation as it is untested
* test build fix
* Make sure reverb samples don't overflow
* Fix up SPU save states
* Wrap negative reverb buffer offsets
* Clear dma req when ADMA is actually finished
* Don't skip voice crest offset 0x1FF
* Disgusting reverb implementation
* nocash style reverb reg names
* Start reverb work, pre-renames
* Write out raw voice1 and 3 data to memory
* EVOL, AVOL regs
* Cast ADMA input before applying volume
* Simplify adpcm decoder
* Move SPU to its own directory
* Update ADSR params on register write.
* Clean up attempt made
* Volume sweep mode, wanna clean this up later
* Shift volume to the right value
* Save MMIX register writes
* This looks slightly cleaner at least
* Implement BVOL, make SPU's read high regs.
* Probably best to initialize these
* Write ADMA to SPU Memory and then Mix it to output
* Actually switch phase properly on ENDX
* Read all the bits of the asdr shift value
* Setting sustain level high enough caused overflow
* Printf on volume sweep mode
* Mute voices if vmix state is off
* Effect area regs
* Write regs PMON and NON.
* General cleanup of adsr stuff
* Turn off ADSR processing after end of release
* Initial hacky ADSR crap
* Gaussian interpolation
* This is not correct and didn't help as a temp hack
* Clean up leftovers
* Mix voices and output wav per SPU
* Tidy up interpolation, split up sample gen method.
* Write out voices in stereo
* Add stereo wav writing
* Remove old wav wav write function.
* Output audio at the right frequency
* Restore voice dumping to working order.
* Also build for msvc
* Add audio files to qmake
* I'm dumb
* Write 48000hz into the wav header
* ps-adpcm decoder with hacky per voice wav output
* Add ability to dump SPU voice data to disk

Download: DobieStation Git (2020/06/10)
Source: Here