Arbee's To-Do List: Difference between revisions

From MAMEDEV Wiki
No edit summary
(12 intermediate revisions by 2 users not shown)
Line 1: Line 1:
Nothing listed here is a guarantee or promise of action.  I have many more interests than I have hours in a day.  This lists only things directly related to MAME.  Also, order listed here may not reflect actual implementation order.
Nothing listed here is a guarantee or promise of action.  I have many more interests than I have hours in a day.  This lists only things directly related to MAME.  Also, order listed here may not reflect actual implementation order.


==Near-term==
* "napple2gs" project
* Finish emulating DEC Rainbow's keyboard, try to boot MS-DOG
  - New clean, modern IIgs driver, running the ADB micro controller code (in progress)
* Line-level ADB handling so Egret/Cuda equipped Macs have at least a mouse working (and IIgs when ADB MCU is properly emulated)
  - Simulate 1 MHz synch slowdown (initially done)
* C++ 6502 cycle-by-cycle core (and adapt the rest of the family to be based on this framework)
  - Working IIe keyboard for simulated "upgraded IIe" (working)
* M50740-series core (6502 derivative) for IIgs ADB MCU / PMU on PowerBook 1xx
  - Pass self-tests and diagnostics disk
* nscsi-based NCR5380 and 5394/96 adaptations
* Emulate key repeat on Apple /// (similar to //e, same h/w designers, go figure)
* Help Andrew with ''Polygonet Commanders'' where possible (I AM NOT A DSP56K COWBOY)
* Break up Mac driver to better reflect h/w differences, and make the various system ASIC families into devices (in progress)
* Continue to work with the SpriteMind guys on YM2612 improvements (if any)
* Figure out Ensoniq 16-bit crash (EPS OS, VFX SD/SD-1 sequencer)
* SH-4 DRC for Naomi/Atomiswave (and Dreamcast in MESS)
* Emulate 3rd-generation Ensoniqs (ASR-10, MR-Rack, KT-76)
 
* Break out HLE'd line-level ADB handling into a bus/ and devices.
==Later==
* Get Lisa to run
* SCSP/AICA DSP recompiler
* SH4 and ARM DRCs
* AICA needs better envelope handling.  Shenmue DSF rips need the filters.
* AICA needs better envelope handling.  Shenmue DSF rips need filters
* 65816 family reunion
* Cycle-by-cycle WDC65816, integrating M377xx and Nintendo variants
** Merge 65816/M37710 so identical opcodes use the same code
** Add Nintendo varient that runs in terms of SNES master cycles for better accuracy
* Rewrite Konami video chain and trojan unclear details
* Further reverse-engineering on (Super) System 23
* Investigate 680x0 instruction restart
* ARM DRC
 
==SDLMAME/MESS==
* Rework OpenGL backend:
** Need to use a pixel shader to do YUV to RGB conversion or else laserdisc games will suck mightily on modern video cards (which no longer have dedicated conversion hardware)
** Render to texture first so user shader can still be used for final blit?  (And how bad will this break fglrx and Apple drivers?)

Revision as of 16:21, 6 August 2021

Nothing listed here is a guarantee or promise of action. I have many more interests than I have hours in a day. This lists only things directly related to MAME. Also, order listed here may not reflect actual implementation order.

  • "napple2gs" project
 - New clean, modern IIgs driver, running the ADB micro controller code (in progress)
 - Simulate 1 MHz synch slowdown (initially done)
 - Working IIe keyboard for simulated "upgraded IIe" (working)
 - Pass self-tests and diagnostics disk
  • Emulate key repeat on Apple /// (similar to //e, same h/w designers, go figure)
  • Break up Mac driver to better reflect h/w differences, and make the various system ASIC families into devices (in progress)
  • Figure out Ensoniq 16-bit crash (EPS OS, VFX SD/SD-1 sequencer)
  • Emulate 3rd-generation Ensoniqs (ASR-10, MR-Rack, KT-76)
  • Break out HLE'd line-level ADB handling into a bus/ and devices.
  • Get Lisa to run
  • SH4 and ARM DRCs
  • AICA needs better envelope handling. Shenmue DSF rips need filters
  • Cycle-by-cycle WDC65816, integrating M377xx and Nintendo variants