Project Logs

Collapse

• Development environment outline

  fjkraan • 12/18/2025 at 14:58 • 0 comments

  The old situation relied on the ROM Emulator to upload code into the MPF-I memory space. With the hex-load routine, this is now possible from the serial monitor. As this upload uses the same serial connection as
  the monitor console, so it is probably not very sound. At the other hand it seems to work and similar
  configurations work for the 6809-BEM3 and SPC16/10 configurations. 

  The potential problem exists at the host side, where the terminal program connected to the serial port must 
  allow another program to feed the new code (in .hex or .s19 format) to the monitor.

  For the MPF-I dev.env, the terminal is a PuTTY session for the commands and the ROM Emulator Terminal for
  the hex-feed. As there is no handshake, a 0.1s delay is inserted between the lines. With the other dev.envs,
  a cat file.hex > /dev/ttyUSB0 works, so there is some handshaking there. To be investigated.

  Re-investigated minicom again, as this should be the integrated Linux terminal & download program. Found 
  a reason why it didn't work; it was checking the modem handshake line DCD. '^A o' > 'Modem and dialing' > 'R'.
  It still displays OFFLINE, but works. In the main configuration menu, select 'Save setup as dfl'. This 
  created a .minirc.dfl file:

  # Machine-generated file - use setup menu in minicom to change parameters.
  pu pname9           YUNYNascii
  pu pprog9           ascii-xfr -dsv -l 100
  pu baudrate         9600
  pu bits             8
  pu parity           N
  pu stopbits         1
  pu updir            ./
  pu downdir          ./
  pu hasdcd           No 
  pu rtscts           No 

   The 'pprog9' option is modified, inserting 100 ms delay between the lines of the hex-intel file.

• Software & hardware changes

  fjkraan • 10/12/2024 at 13:17 • 0 comments

  Some of the recent changes:
  * The monitor now can use either the DART Channel A or B, depending on a single I/O-port address,
  * A four pass RAM test is introduced, which executes seven checks on each address. This will catch most RAM chip and adressing errors,
  * For the MPF-I, most keys on the keyboard can be tested,
  * On the hardware front, a new version of the board is on its way, using
  a 16x BAUD rate clock. This should improve the (still planned) Hex-Intel transfer command.

View all 2 project logs