The Sinclair QL Forum

Have managed to serial transfer the "errmsg.proc" to my laptop (and add to ROM disk image) and instead of hanging Q-emuLator it now gives proper error descriptions as it should, which suggests the transfer was good

Later today I'll try and transfer all the other programs on the 68K/OS mdv, so we have working copies, and then install and test them into the ROM disk image

t0nyt wrote: ↑Mon May 06, 2024 10:39 am Have managed to serial transfer the "errmsg.proc" to my laptop (and add to ROM disk image) and instead of hanging Q-emuLator it now gives proper error descriptions as it should, which suggests the transfer was good

Later today I'll try and transfer all the other programs on the 68K/OS mdv, so we have working copies, and then install and test them into the ROM disk image

Awesome, your doing a fantastic job there!

Ok, so attached is a new ZIP containing the 17 files that are on the 68K/OS Cart. Maybe this can be added to Dilwyns download page archive for 68K/OS?

These are all tested (other than COLOUR, DRAW & SLIDES, but they should be ok), as far as it's possible to test from ROM, in Q-emuLator

So, as far as I can tell, they are all good (TIME_PROG which uses CLOCK_PROC hangs though, but I think this may be a problem with running it from ROM as it strangely has a double header, but will investigate further as time allows. So in the ROM I made CLOCK_PROC executable)

I'll attach the ROM in a separate post when I finish a small document explaining how it's structured

Ok, this is the ROM bin file which can be added as the "Back" ROM in Q-emuLator

I've only tested this in Q-emuLator for Mac so far (EDIT: also works in Windows Q-emuLator)

It contains 14 of the 17 files (excludes COLOUR, DRAW & SLIDES)

Also attached is a description of the ROM layout for posterity (will update if I find out anything else)

Once I've replaced my MDV pinch rollers, and re-felted the cartridges, I'll try to preserve the contents of the 68K/ASM carts (if the master or backup is still good)

t0nyt wrote: ↑Sun May 05, 2024 7:52 am This is certainly something I'd be keen to do when circumstances allow (I'm a 24/7 carer with little time or space to do something like this), but if you can let me know exactly what I should get I'll get it in ready so I can give it a go as circumstances allow.

Hi Tony,

The logic analyzer I have is by HiLetGo, but I see there are others that look exactly the same and cost even less.
I would recommend PulseView (free) for the software on the Mac/PC side, as it saves the traces in a format that I am able to read.

The main issue is how to attach the wires from the analyzer to the microdrive connector. Here are some ideas, but maybe someone better versed in electronics than me can offer additional ones:
- Connect a MDV cable if you have it, and use probe hooks between the analyzer and the cable
- Buy a connector for the MDV PCB connector and attach the analyzer to it with the cables that come with the analyzer
- Use hooks to connect to the lines inside the QL (e.g. microdrive ULA pins)

The process to collect the trace should be pretty easy:
- re-felt the MDVs and connect the logic analyzer to the two data lines and GND (on the side microdrive connector, or inside the QL)
- boot the QL (standard Sinclair or Minerva ROM is fine), insert microdrive and turn the motor on (I can send you code for that if you don't have it already)
- on the Mac or PC, run PulseView and start recording at the max frequency (24 MHz), say for 1G samples, that should be about 40 seconds and allow recording the entire tape at least five times
- email me the output for decoding and creating a microdrive image (in non-standard format... from memory, I think they used 100 sectors of 1 KB each).

Parts needed:
- logic analyzer
- either a PCB connector, or hooks for the logic analyzer with optional MDV cable
Later I can send you pictures of the various components I mentioned.

Thanks for the software preservation work you are doing!

Daniele

Hi Daniele,

Have ordered a logic analyser and hooks, will let you know when they arrive

Thanks
Tony

t0nyt wrote: ↑Mon May 06, 2024 8:42 pm Hi Daniele,

Have ordered a logic analyser and hooks, will let you know when they arrive

Thanks
Tony

HI Daniele,

I now have these items:

Attached is a zip of the files on the 68K/ASM (2 are zero bytes, but they are the same on the original master and I don't think they are important)

I got excited because it includes a "DISK:" device driver, which I can load on the real 68K/OS, giving me the DISK: device but obviously I can't connect my Trump at the same time to see what happens

The reason I got a bit excited was I thought I could add it to the ROM image and maybe access floppy images in Q-emuLator BUT the file is "text only" which means it has NO HEADER to allow it to be added to the ROM image. Maybe there IS a way, but I can't see it at the moment (tried a header pointing to the text location in ROM but that didn't work)

EDIT: the 2 zero byte files are just work files. SPACE.TOP is a work file from EDIT.PROG, SPACE.BIN is part of the assembler output.

The DISK.DRIVER file is quite interesting as it is assembler source code and if compiled would have a file header (the first few lines define the header)

But 68K/OS (or ADAM) does load it and make the "DISK:" device available, which would suggest ADAM can compile assembler on the fly?

When I have a chance I'll try to compile it on the real hardware and see if that's loadable via the ROM

EDIT: I notice in the driver file at one place it says "Do" when it should say "D0"

EDIT 2: Also, I can't find any definition for "INSTALL" (it's not in the driver file nor either include file)

EDIT 3: this entry in the manual suggests it should be able to put the file into ROM:

"It is possible to add device drivers to the system. This is usually done by running a program which uses special 68K/OS system calls to install the driver software. If on start-up 68K/OS finds any programs in the ROM: directory which have .DRIVER as an extension these are run immediately: by this means drivers may be loaded automatically."

If it's true, I wish I could work out how!

EDIT 4: it doesn't compile any better under 68K/OS than under QDOS GST Assembler even if you include the _IN files into the source with either