Difference between revisions of "Todo"

From vice-emu
Jump to: navigation, search
m (new ports)
m (xcbm2)
Line 413: Line 413:
* cbm2 z80 support.
* cbm2 z80 support.
* cbm2 8088 support.
* cbm2 8088 support. ([http://www.cbm-ii.com])

Revision as of 01:30, 6 March 2011

Prime Directive

  • when adding/changing stuff that needs (G)UI items/changes, add a proper entry to the Developer_log to notify the UI maintainers.
  • when adding/changing command-line options and/or resources, update the documentation.
  • after fixing a bug, update its status on the Bug Tracker and mark it as fixed in the list below.

Known Issues

This is the list of known problems. also look at Tracker: open Bugs

This list always refers to the status of the last major release version, which was 2.3. Items that have been fixed by the time of a major release will be removed, items fixed after that will be marked as fixed and stay until the next major version. New items will be added as noticed.


  • emulator system state is always the same when starting, meaning that various methods of randomizing data will fail on vice (ie, always give the same result) when using the autostart feature.

Workaround: load the program in question manually, that will usually introduce enough randomness into the system state.

note: this could perhaps be solved (atleast partially) by inserting a random delay before "run" ? --Gpz 01:45, 11 January 2011 (UTC)


  • the documentation is partly outdated and/or incomplete.


  • When a key which is shifted on the real machine but unshifted on the PC or Unix keyboard you are using is pressed, the virtual shift is pressed together (i.e. at the same clock cycle) with the main key; this could not work with some keyboard routines.

note: also see this knowledgebase article: http://vicekb.trikaliotis.net/20-006

Workaround: If you have to use the function keys, press the (real) shift key manually (e.g. F2 = Shift + F1, F4 = Shift + F3 and so on); this also works with other keys.


History Recording

note: since history recording is very error prone and hard to debug, we depend on feedback (on either failure or success) on this topic very much. if you still have problems with the current release, please report them including testcases.

CRT Emulation


note: This is not really a bug. The result of the opcode in question is A = ((A | CONST) & X & IMM), with CONST apparently being both chip- and temperature dependent. In other words: this particular loader will also not work on all real C64s. (see testprogs/general/ane-lax)


  • the handling of MCBASE at power-on is incorrect (see testprogs/VICII/spritemcbase)

note: this should only be visible directly after power-cycle on a 6569


  • the case when CIA1 port A and port B are both in output mode, port A outputs low, port B outputs high, both are connected via the keyboard matrix and a value is read from port B is not handled correctly ( testprogs/CIA/ciaports ; http://noname.c64.org/csdb/release/?id=5375 ) (improved in 23609)

note: the current implementation is correct for single key presses (port B reads 1), which should be enough for most practical situations (and to make broken keyboard routines, such as the one in nitro#16, work). the following special cases (which would make port B read 0) are not emulated yet:

  • shift-lock is pressed
  • two ore more keys of the same column are pressed


  • the POTX/POTY sampling period is not taken into account
  • POTX/POTY sampling jitter is not emulated
  • only one mouse or pair of paddles can be emulated at a time


Drive Emulation

  • Dual disk drive 2040/3040 support (DOS1, 670 blocks free) is not working
  • when using a dual drive, only the first drive (0:) can be used.

virtual Devices

  • the drive rom can not be read using "m-r" when tde is disabled (instead zeros will be returned).

note: this is actually not a missing feature but a bug - because the code _does_ return the drive rom array, its just not loaded. however, the current situation (returning zeros) should be default for most practical situations - so when this gets fixed, it should be optional. --Gpz 01:52, 11 January 2011 (UTC)

True Drive Emulation

x64 / x64sc

x64 issues addressed by x64sc

To clarify... these will never (as in can't) be emulated in the regular x64, unless we end up replacing x64 with x64sc.

  • some VIC-II bugs present in the old VIC-II core
    • inline gfx data changes are not emulated (see testprogs/VICII/gfxfetch/)
    • The VIC-II implementation lacks cycle exact sprite collision support
    • VIC-II shows wrong colors when switching from hires background to idle with borders open (see testprogs/VICII/border/*, Krestology)
    • VIC-II fetches from cartridge in ultimax mode can not be emulated properly for all carts

note: this is also broken in xcbm2

note: this is also broken in xplus4, xcbm, x128, xpet, ie most emulators that do not use the new cpu core (works on x64dtv due to special zeropage handling)

Workaround in all cases: use x64sc :)

x64 specific problems

note: While most of the x64sc improvements/fixes are impossible to shoehorn into x64 (hence the split), a few things could still be improved:

  • testprogs/VICII/spritesteal (ntscold)
  • testprogs/VICII/phi1timing (ntsc,ntscold)
  • testprogs/VICII/spriteenable
  • missing open border on top of testprogs/VICII/sprite0move
  • testprogs/VICII/split-tests/modesplit

Workaround in all cases: use x64sc :)

Cartridge System

  • some carts have no snapshot support yet (MMC64, Magic Voice, MMC Replay, IDE64, MIDI, SFX Sound Sampler, SFX Sound Expander, Ethernet)
  • some carts do not work 100% yet (Magic Voice, MMC Replay, Stardos, Snapshot 64)
    • Magic Voice: works only in basic mode right now
    • MMC Replay: only the BIOS mode works right now; the monitor may incorrectly access cartridge memory (even) if "bank cart" is explicitly set
    • Stardos: Disk copier does not work for some strange reason
    • Snapshot 64: "format and backup" (freeze, then F5) hangs after entering name of boot file
  • IDE64 shortbus expansions and cdrom/iso support is not emulated
  • SFX sound sampler audio input is not emulated

note: the detailed todo list is here.


  • Support for fast mode (2MHz) is implemented but incomplete, detailed information about the stealing of cycles by the VICII when switching back from fast mode is missing and therefor the emulation of it is incorrect.
  • break- and watchpoints in zeropage are broken. this is because of the old cpu core, and can likely not be fixed without migrating to the new one.
  • support for c128 specific expansion port handling/mapping is missing in the c64/c128 cartridge system (MMC Replay, FastLoad128, MACH128, other?)
  • if CRT emulation is enabled, it is always enabled for both the VIC-II and the VDC output, and both share the same settings.

note: to fix this all "ColorXYZ" and "PalXYZ" resources and respective command-line options must be changed to be per video chip (as with other settings). at this point it is probably also a good idea to merge the "CHIPScale2x" and "PalEmulation" (boolean) resources and options into one "CHIPRenderer" enum. Gpz 15:29, 2 February 2011 (UTC)


  • SID card only supports "fast SID"
  • NTSC bugs:
  • Idle fetch (or the lack of it) needs investigation
  • "Screen mixup" is not emulated
  • COLS=0 can lead to negative width in raster/ causing segfaults if not checked for in archdep video code




  • Although both are emulated in the xcbm2 emulator, switching between C510 and C6x0/7x0 is not possible at runtime.

Workaround: Use command line option "xcbm -model 510" or "xcbm -model 610" respectivly.

note: this will be solved by splitting the cbm2 emulator into two binaries

  • The CBM-II emulator, when the execution bank is set to an open memory bank, sets the zeropage and stack access to unmapped, but actually existing memory. This is a bug but cannot be avoided with the current CPU code architecture.
  • break- and watchpoints in zeropage are broken. this is because of the old cpu core, and can likely not be fixed without migrating to the new one.
  • banking in the monitor is broken


  • In the C510 emulation the VIC-II timing is not exact, i.e. the stopping of the CPU during bad lines is wrong.
  • The C510 emulation uses the old VIC-II core which means no cycle exact sprite collisions, no inline gfx changes, and probably more.


note: run like this: xcbm2 -model 610 -ntsc -drive8type 8050 +truedrive -8 cbm2-burnin.d80


  • When running the "8296d diagnostic" test of the 8296 system disk on "xpet -model 8296" some tests fail:
    • when TDE is enabled, the program will hang. when TDE is disabled, the IEEE interface status will be reported as bad
    • the userport is reported as bad (perhaps some dongle is needed?)

note: run like this: "xpet -model 8296 -drive8type 8050 -8 8296d-systemdisk.d82 -editor edit-4-80-b-50Hz.901474-04.bin". the "burnin8296" program seems to be an older version of the "8296d diagnostic" program. it runs when TDE is enabled, but the 50Hz irq test fails.

  • break- and watchpoints in zeropage are broken. this is because of the old cpu core, and can likely not be fixed without migrating to the new one.
  • some combinations of selected (CRTC) model and "double size" mode give not the correct/expected result.
  • some combinations of selected keymap and keyboard model do not work because of missing keymaps

Ports / UIs

  • Screenshots with activated PAL emulation do not deliver the expected (blurry) result.
  • In the C128 emulation VIC-II and VDC palette can not coexist currently in some ports.
  • The CBM-II emulation is not fully supported in all ports.


Win32 DX9
  • There is a bug when moving the window partly outside the left edge of the screen. The rendered contents seem to "lock" to the left border. This seems to depend on the driver of the gfx adapter as is only shows on some machines.
Windows 7

X11 (XAW and GTK)

  • The custom TextField widget is buggy and can cause crashes on DEC/Alpha machines, and possibly on other systems too. If you get weird X protocol errors or the file selector makes the emulator die when you close it, this might be the cause.

Workaround: recompile with the `--disable-textfield' option. If it still does not work, please tell us.

  • Aspect ratio/scaling is broken in fullscreen mode when hardware rendering is enabled (canvas is always stretched to fill the entire screen)
  • the mouse pointer is incorrectly un-grabbed when switching from/to fullscreen with the mouse- or lightpen emulation enabled.
  • lightpen coordinates are not accurate when using vidmode for fullscreen
  • Gnome/GTK+ commandline is ignored.
  • Some DEC/Alpha machines crash due to the XAW implementation.

Workaround: Compile VICE with GNOME support `--enable-gnomeui' instead.

  • Context popup menu for drive attach is broken if more than two drives are active.


  • In the MS-DOS port you have to save the configuration and restart the emulator when changing VDC double size mode.




  • Sometimes the application loses the keyboard focus; no keystrokes are accepted inside the emu, so the emulator has to be restarted.
  • Sound problems on slower machines (P100) with a non-official driver for sb16.
  • The monitor only works if you start the emulator from a terminal; if you start the emu from tracker and enter the monitor, you are "lost"
  • Fullscreen mode is not implemented.


  • The GP2X port segfaults at start.

note: this port is unmaintained and because of that subject for removal.


  • uses preset sdl-vicerc files which are unmaintained and thus could fail. At least sdl-vicerc-cbm5x0 is obsolete and broken due to the config split.

note: this port is unmaintained and because of that subject for removal.

New Features

This is the place for the VICE developers to put down ideas about the future development of VICE.

Some of these ideas might never be realized, and might look completely absurd, but they could be the inspiration for other ideas.

By putting these ideas down they will be known to any other developers that would like to pick up one of the ideas and implement it.

also look at Tracker: open Feature Requests


Sound System

Video System

  • some kind of generic interface to handle video input is required to handle scanner/digitizer type of carts

Hardware emulation


  • internal (?)
    • c64/c128 1.44 MB disk drive support. ([1])
    • c64/c128 SCSI expansion support. ([2])
  • Cartridges
    • IDE64 improvements
      • IDE64 cdrom/iso support.
      • IDE64 eth64 shortbus expansion support. ([3])
      • IDE64 duart shortbus expansion support. ([4])
    • SFX sound sampler improvements
      • SFX sound sampler audio input support. ([5])
    • Clockport improvements
      • Silversurfer (UART) support ([6])
      • mp3@c64 support
    • Community Information Locator Commodore 64/128 addon cartridge ([7])
    • Pagefox cartridge ([8])
    • RAM Floppy (REX) support
    • Daisy Sound sampler audio input support. ([9])
    • PS64 speech cartridge support. ([10])
    • Voice Messenger speech cartridge support. ([11])
    • ramlink support. ([12] [13])
    • ramdrive support.
    • rtc (ds12c887) support. ([14]) (in progress).
    • z80 (cpm) cartridge support. (4Mhz z80)
    • turbo master cartridge support (4Mhz 65C02). ([15] [16])
    • turbo process cartridge support (4Mhz 658156). ([17] [18])
    • flash-8 cartridge support (8Mhz 65816). ([19])
    • SCPU cartridge support (20Mhz 65816). ([20] [21]) (https://sourceforge.net/tracker/?func=detail&aid=2030692&group_id=223021&atid=1057620)
    • The Final Chesscard (65c02) ([22])
    • german BTX cartridge (6809)
    • SoftPROM cartridge ([23])
      • somewhat pointless imho, unless software specifically designed for this cart can be found (which i doubt). other than that its just an ordinary 8k game cart (with battery backed up ram instead of rom). Gpz 23:01, 30 October 2010 (UTC)
    • Net64 cartridge support. ([24])
      • this is exactly the same as "the final ethernet" (which is emulated) Gpz 23:27, 30 October 2010 (UTC)
    • MasC=uerade Cartridge Converter support, this cart is originally intended to use c64 carts on the vic20, but in theory could also be used the other way around. ([25])
    • BI-80 cartridge support. ([26])
    • Buscard II IEEE488/printer-port cartridge support. ([27])
  • Userport
    • C64 RTC 58321 (userport) ([28])
    • CBM1660 (userport) modem300 support. ([29])
    • CBM1670 (userport) modem1200 support. ([30])
    • FB-RS232 (userport) support. ([31])
    • c64 p64 midi interface: ([32])
    • Scantroniks handscanner (userport) support. ([33])
    • Video Byte II video digitizer (userport) support. ([34]) ([35])
    • Computera Eyes video digitizer (userport) support. ([36] [37])
    • Userport RocketDrive support. ([38])
    • CIA-IDE-HD support. ([39])
    • Aprotek universal rs232 adapter support. ([40])
  • Joystick Port
    • BB-RTC (joystick port 2) support. ([41])
    • C64 smartmouse RTC support.
    • c64/c128 numeric keypad support. ([42])
    • c64/c128 pc-joystick support. ([43])
    • datel 3-button mouse support. ([44])
    • Koala-pad support. ([45])
    • Covox Sound Master audio input (joystick port) support. ([46])
    • animation station support. ([47])
    • Power-pad support.
    • super sketch support. ([48])
  • Video output.
    • c64 LCD64 screen support. ([50])


  • c128 plus60k support.
  • c128 plus256k support.
  • c128 256K memory expansion hack support.


  • VIC20 RTC cartridge support. ([51])
  • vic20 iec2ata support. ([52])
    • pointless to emulate, as it is a simple iec device, very similar to what you get when using virtual drive on the filesystem. Gpz 05:08, 15 August 2010 (UTC)
  • vic20 ethernet support. ([53])
  • MasC=uerade Cartridge Converter support. ([54])


  • plus4 digiblaster audio input support. ([55])



  • cbm2 z80 support.
  • cbm2 8088 support. ([61])



  • IEC
    • DPS1101 support.
    • MPS801 support.
    • MPS802 support.
    • MPS1000 support.
    • MPS1250 support.
    • Fidelity Impact Printer support. ([72])
  • IEEE
    • MPS1361 support.
    • CBM1520 support.
    • CBM1526 support.
    • CBM4022 support.
    • CBM8023 support.
    • CBM8024 support.

new Ports/GUI improvements

  • Make a list of GUI elements/features per port, compare and implement any missing features for the ports.
  • Debug/fix the XBOX SDL port.
  • Debug/fix the RiscOS SDL port.
  • Debug/fix the Windows-CE SDL ports.
  • Completely update the RiscOS GUI.
  • Completely update the GP2X GUI.


  • add "commandline options" help window (like in the windows port)
  • add Drag'n'Drop support
  • add Session Management
  • improve Tape status widget, (.TAP vs .t64) handling is lousy.
  • implement Monitor GUI
  • add VIC20 Cartridge attach heuristic for image pairs


  • add "commandline options" help window (like in the windows port)


  • add PCI support for catweasel and hardsid


  • add PCI support for catweasel and hardsid
  • add real IEC device support via par port or ser port
  • add proper lfn detection

new ports

  • windows-ce X11 support.
  • amiga X11 support.
  • RiscOS X11 support.
  • Port to Plan9
  • Port to win16
  • Port to win32s (maybe by using windib)
  • Port to macos classic (8.x/9.x).
  • Port to VersaDOS
  • Make OS/2 SDL port.
  • Make atari mint SDL port.
  • Make macos classic SDL port.
  • Make dreamcast SDL port.
  • Make GP32 SDL port.
  • Make PSP SDL port.
  • Make WII SDL port.
  • Make PALMOS SDL port.
  • Make GBA SDL port.
  • Make NDS SDL port.
  • Make PS2 SDL port.
  • Make GameCube SDL port.
  • Make VAX OpenVMS SDL port.
  • Make ALPHA OpenVMS SDL port.
  • Make HPPA OpenVMS SDL port.
  • Make Symbian OS SDL port.
  • Make XBOX360 SDL port.
  • Make Ipod/Iphone SDL port.
  • Make android SDL port.
  • Make pandora SDL port.
  • Make caanoo SDL port.
  • Make wiz SDL port.
  • Make GP2X SDL port with proper joystick/pad handling.
  • Make zodiac SDL port.


  • Split xcbm2 into a 5x0 binary and 6x0/7x0 binary.
  • Split the vsid mode off from x64, making vsid a seperate binary.
  • The "sc" rewrites. Migrate emulators to the cycle based CPU core (6510dtvcore.c, should be renamed), which allows:
    • cycle based hooks for complex hardware expansions (SCPU)
    • in-line graphics data change emulation (see testprogs/VICII/gfxfetch)
    • cycle exact Blitter/DMA vblank start and proper LinearA/B counter handling on x64dtv
    • cycle based drawing
  • ...and requires (at least):
    • rewrite video chip modules to use:
      • cycle based fetch/etc (VIC & VICII done, VICII-DTV in progress)
      • cycle based drawing (VICII done)
    • remove 1 clock write offset in CIA etc (at least CIA/VIA already handled in trunk)
    • an insane amount of regression testing (preferably with new testprogs where needed)


  • Update the *nix documentation
    • generate info.c from vice.texi
    • perhaps find a way to conditionally include arch specific stuff in vice.texi ?
  • proper and specific win32 documentation. (.chm format)
  • proper and specific amiga documentation. (.guide format)
  • proper and specific BeOS documentation.
  • proper and specific MSDOS documentation. (.txt format)
  • proper and specific OS/2 documentation.
  • proper and specific RiscOS documentation.
  • proper and specific SDL documentation.


  • Support for more languages.
  • Translation support for BeOS.
  • Translation support for MSDOS
  • Translation support for OS/2
  • Translation support for RiscOS.