DigicoolThings Forum

Wow, it looks as if you have made my wish come true. Ive been wanting, meaning to get around, to making a flex system. and you've just handed it to me on a plate. must get back to doing some MECB stuff soon now.

(ive been working on some old bus station led matrix displays for last few weeks - using a new firmware using rp2040w and some of the old drivers instead off the old z80 where sadly the original firmware lost. just about sorted now ready for demo installation in the museum)

I'd have admit that it has only been lightly tested but thought it best to post whatever I have so far (before I get distracted) in case it is useful to others (LOL, in helping their wishes come true). I managed to get trek and Othello running on the DigiBug 6800/2 system so posted those as well (though those run in the monitor and not under the OS - I haven't really looked at what is needed to cross-compile to FLEX but that might be an interest thing to do - sigh, so many things to do). I should also look into what is needed to get NitrOS9 running (I suspect that might be more involved - but it is a much nicer OS, one that was really ahead of its time).

The LED matrix sounds like fun - I look forward to seeing it in action

I found a kind of work-around for cross-compiling - I can compile on my main computer using asl and then in FLEX type "MON" to enter Digibug, load the S1-format output, do a warm start of FLEX "G CD03", and then a "save FILENAME,HEX-BEGIN,HEX-END,HEX-EXEC" - that saves it on to the FLEX disk. I tested this by modifying the FLEX DATE command to now read the network time off the FujiNet and it works a treat (it's even better than the real-time clock because you don't ever need to set it or worry that the battery will run out ... though an Internet connection is required)

After quite a bit of messing about I managed to get NitrOS9 booting off FujiNet (and fetching the network time).
I got thrown off on a bit of a wild goose chase with something I still don't quite understand - the FujiNet library was a bit too big to fit in the OS9 kernel with a 4 KB ROM so I tried to extend the ROM size to 6 KB (i.e. slight increase to leave as much RAM available as possible) but that was having some peculiar effects on the bus (I/O was kind of working but not quite e.g. I could start the timer but when I read the status register it was always $ff). I eventually gave up and switched to 8 KB ROM and 56 KB RAM and that seemed to work - unfortunately that means the system now has less memory to play with.

The screen shot above has the booting sequence where it reads off the FujiNet-defined disk and reads the network time. I then log in, compile an assembly program, and run the binary as a background task (it creates a Cylon-like effect on the IO port LEDs).

Now working my way up to the 68008 board. Ported across the Fujinet library to work with it, wrote a very basic BIOS, and was able to boot up CPM 68K v1.3.
It took ages to get to this point - mainly because the information available for this OS on the Internet is a complete dog's breakfast with a mix of versions, old system guides and binaries that require patching to run. But anyway, finally got to a point where it appears to be fully functional - tested this by writing a hello world program in c (using the CP/M editor ed), compiled it, linked it and ran it.

Glad to tick this off my bucket list

I have the fujinet library now ported across to the 8088 board:
Next challenge is to see if I can get some OS ported across to work with this (it's hard to find anything that can run on a generic 8088 that doesn't look and behave like an IBM PC).

Another OS ticked off the bucket list - CPM-86 ... now working with the Fujinet on the MECB 8088 board:
The only board without an OS now is the 6502 ... I'd have to admit it is my least favourite CPU to play with. Perhaps once I tidy up a few loose ends with CPM 86 and maybe check wether if it is at all possible to get DOS running on a non-IBM 8088 computer ... I'll see if I can finish off this journey with an OS on the 6502 board.