dml attempts a Raven

[dml]

Could the chips not be programmed in a normal programmer?

These socketed ones can for this basic board but the image needs sliced and diced to program them separately. I was recommended just to finish this board instead. Also - I built the more fancy version of the ROM board with the larger capacity SMT flash chips which really needs this programmer anyway.

[dml]

So I upped the stakes from 'weekend rummage' to 'full emptying of everything at all costs, no matter how much tidying needed afterwards'.

And I found exactly x2 HC273s.

two.jpg (610.27 KiB) Viewed 394 times

Not optimal given the 3rd one is a 74LS - but they don't talk to each other, correct? They're parallel so if nobody says anything they won't know any better? :p

It's this kind of thing that has me destroying components vs being patient.

[exxos]

It be fine

[dml]

It be fine

I decided the same now its plugged into a machine, lights are happening and I'm trying to figure out what to do!

[dml]

So I made a bit of progress on this now but I think I reached a point where I need to go back and read 300 forum posts to get the next step.

- Installed the HC273s
- Grabbed the raven git repo
- Grabbed the pico-sdk git repo
- Configured/built the rom programmer subproject
- Plugged the programmer Pi into the Mac
- Copied the .uf2 to the Pi, watched it reboot. The very bright green LED is now on. I hope that's good.
- Got CoolTerm, configured it for 230k baud
- Connected to the programmer......

.....

not seeing anything.

Checking the programmer readme/page it says 'Usage: Connect with a serial terminal and follow instruction.'.

I'm connected I think, and I can't locate or see any instruction. This is the point where reading a lot may help :p

[dml]

Doh. well that didn't take long. I'll just say that it's working now after a slight derp.

Screenshot 2025-06-28 at 17.58.35.png (73.35 KiB) Viewed 377 times

Ok I got slightly confused by the programmer going into nonresponsive mode after issuing a command - but it was just the CR+LF config in the terminal. Fixed now.

Code: Select all

> i
[$bfbf:$d7d7] SST : SST39LF/VF040 

So I seem to be getting reliable responses from the programmer. Time to try flashing something.

[agranlund]

- Got CoolTerm, configured it for 230k baud

You can go faster than that but I have no idea at what point that would become pointless due to programming speed, I've always just connected with 750k baud.

[dml]

You can go faster than that but I have no idea at what point that would become pointless due to programming speed, I've always just connected with 750k baud.

Is it wise to start with the monitor ROM or go straight to the full ROM? Start with the monitor perhaps?

I'm assuming there are going to be a bunch of things wrong with my build :p

[dml]

Have been struggling a bit recently to find time to work on things but this morning have made a bit of progress with the programmer.

Managed to flash the monitor ROM and confirm a dump of random fragments against the binary file.

prog.jpg (330.27 KiB) Viewed 294 times

So at least I know I built the programmer correctly even if it wasn't the most challenging part.


Probably I can try with the SMD ROM board next to confirm that is ok since that was a bit more tricky...

[dml]

Is there a good way to verify the contents of the whole ROM? Does the programmer do this itself after writing or does it require using the [d] dump operation and capturing the stream? It appears to be pretty-printed with addresses so verification with that might be tricky.

In any case, it appears the SMD ROM board has been written with the full TOS image so assuming the write was successful, I should have a monitor ROM and a TOS ROM to switch between while I try to bring the machine up.