Loading ...
Sorry, an error occurred while loading the content.

RoboJDE w/thirdparty AVR boards

Expand Messages
  • jeffskubick
    How closely does a thirdparty AVR-based controller need to mimic the Intellibrain in order for a purchased copy of RoboJDE to meaningfully run on it
    Message 1 of 2 , Mar 4, 2006
    • 0 Attachment
      How closely does a thirdparty AVR-based controller need to mimic the
      Intellibrain in order for a purchased copy of RoboJDE to meaningfully
      run on it ("meaningfully" = all the base language classes work, but
      not necessarily the supporting classes for things like sensors,
      motors, etc. in com.ridgesoft.intellibrain.*)?

      Does the chip HAVE to specifically be a Mega128, or does it also
      support the Mega2560/2561? At the other extreme, how about a
      Mega162? (as far as I know, the biggest XRAM-supporting AVR available
      as DIP...)

      Likewise, how much XRAM does it HAVE to have for RoboJDE to work?
      Does the VM blindly assume it has 128mb in the exact configuration of
      the Intellibrain, or can it be configured to work with 64 and/or 32mb
      of XRAM too (either by compile-time settings, or maybe by the VM
      sniffing its runtime environment). If it HAS to have 128mb, is the
      circuit documented somewhere so a RoboJDE-compatible board could be
      designed to work the same way (to get 128mb to work, Ridgesoft would
      have had to implement their own bankswitching scheme, probably using
      one of the i/o registers as the last control line to select the
      desired bank since the Mega128 only supports 64mb of XRAM directly).

      At the low end, support for a Mega162 w/32mb XRAM would be great,
      because it would provide a cheap, breadboard-able entry level platform
      where someone could experiment with RoboJDE Lite before licensing the
      full version or buying an Intellibrain outright (assuming 16mb flash
      is big enough for the VM, of course). At the high end, support for the
      Mega2560/2561 would be easy to add since it's almost identical to the
      Mega128, and would satisfy people who just aren't happy running on
      anything but hardware they designed themselves from the circuit board up).

      In both cases, it would be an opportunity to promote RoboJDE as a
      mainstream alternative to C++ for programming on high-end AVRs. It
      would also bring lots of new developers into the fold, some of
      whom would probably help expand the range of supported hardware to
      include ethernet, zigbee, and more as users wrote native drivers (or
      approximated them using VM.readXXX() and VM.writeXXX()) for more
      exotic peripherals.

      Anyway,as you can probably tell, I'm impressed by RoboJDE, and would
      love to be able to experiment with it for a while before committing to
      a major & fairly expensive hardware purchase (or licensing the full
      version and spending a week or two building my own 2561-based system).

      Thanks!
    • RidgeSoft
      RoboJDE supports the following controllers: - IntelliBrain (www.ridgesoft.com) - Handy Board (www.handyboard.com) - Sumo11 (www.1socrc.com) We do not recommend
      Message 2 of 2 , Mar 4, 2006
      • 0 Attachment
        RoboJDE supports the following controllers:

        - IntelliBrain (www.ridgesoft.com)
        - Handy Board (www.handyboard.com)
        - Sumo11 (www.1socrc.com)

        We do not recommend attempting to use RoboJDE with any other board.
        In all three cases the virtual machine depends heavily on the the
        processor and specifics of the circuit board. If you are looking for
        the least expensive way to try out RoboJDE, used Handy Boards are
        often available at low prices on eBay.

        RidgeSoft Support

        --- In intellibrain@yahoogroups.com, "jeffskubick" <jeff-yahoo@...>
        wrote:
        >
        > How closely does a thirdparty AVR-based controller need to mimic the
        > Intellibrain in order for a purchased copy of RoboJDE to
        meaningfully
        > run on it ("meaningfully" = all the base language classes work, but
        > not necessarily the supporting classes for things like sensors,
        > motors, etc. in com.ridgesoft.intellibrain.*)?
        >
        > Does the chip HAVE to specifically be a Mega128, or does it also
        > support the Mega2560/2561? At the other extreme, how about a
        > Mega162? (as far as I know, the biggest XRAM-supporting AVR
        available
        > as DIP...)
        >
        > Likewise, how much XRAM does it HAVE to have for RoboJDE to work?
        > Does the VM blindly assume it has 128mb in the exact configuration
        of
        > the Intellibrain, or can it be configured to work with 64 and/or
        32mb
        > of XRAM too (either by compile-time settings, or maybe by the VM
        > sniffing its runtime environment). If it HAS to have 128mb, is the
        > circuit documented somewhere so a RoboJDE-compatible board could be
        > designed to work the same way (to get 128mb to work, Ridgesoft would
        > have had to implement their own bankswitching scheme, probably using
        > one of the i/o registers as the last control line to select the
        > desired bank since the Mega128 only supports 64mb of XRAM directly).
        >
        > At the low end, support for a Mega162 w/32mb XRAM would be great,
        > because it would provide a cheap, breadboard-able entry level
        platform
        > where someone could experiment with RoboJDE Lite before licensing
        the
        > full version or buying an Intellibrain outright (assuming 16mb flash
        > is big enough for the VM, of course). At the high end, support for
        the
        > Mega2560/2561 would be easy to add since it's almost identical to
        the
        > Mega128, and would satisfy people who just aren't happy running on
        > anything but hardware they designed themselves from the circuit
        board up).
        >
        > In both cases, it would be an opportunity to promote RoboJDE as a
        > mainstream alternative to C++ for programming on high-end AVRs. It
        > would also bring lots of new developers into the fold, some of
        > whom would probably help expand the range of supported hardware to
        > include ethernet, zigbee, and more as users wrote native drivers (or
        > approximated them using VM.readXXX() and VM.writeXXX()) for more
        > exotic peripherals.
        >
        > Anyway,as you can probably tell, I'm impressed by RoboJDE, and would
        > love to be able to experiment with it for a while before committing
        to
        > a major & fairly expensive hardware purchase (or licensing the full
        > version and spending a week or two building my own 2561-based
        system).
        >
        > Thanks!
        >
      Your message has been successfully submitted and would be delivered to recipients shortly.