• Please review our updated Terms and Rules here

Microkit 8/16

It looks like the motherboard has +5v, -5v, +12 and -12, so I'm guessing the board would have a regulator to drop to -9v. I will explore ot once it is here.
 
I was merging the screen display from the other 'half' of the EPROM - but I am not making much sense of it...

Can you just confirm the wiring of the adapter board that you made up please?

There may be potentially two things wrong:

1. I made an assumption about which data line is MSB and LSB.
2. The wiring is not quite the same as I specified.

The other possibility is that the data you have sent is correct - and my initial partial disassembly was incorrect...

It would be good to find out which before I go any further...

Dave
 
I am resurrecting this thread as I have fully tested my 1702A board for my digital group system, and have now attached the Microkit 1702A EPROM in an attempt to 'dump' it. I placed it in the socket that was marked FC00 and dumped from there:

https://drive.google.com/file/d/1_KrJmN__-8wIYBZUtcA6lCgawWfIVVyi/view?usp=sharing
https://drive.google.com/file/d/1_CNT8Hl9Tq5L7i1VLu7XwptcBoYL23Ee/view?usp=sharing

The 'scan' sort of dimmed the last two lines, but yeah -- hopefully that's useful information. I've no idea how to decode it though.
 
I just went back to your post of EPROM data you were able to decode using the 'video card dump' Dave, and it looks like it lines up! looks like we can fill in the blanks finally!
 
I have a 'raw' disassembly now. It looks good. I compared it to the hex dump we had previously, and only one (1) byte stood out as different. I tracked this down to a fault in my spreadsheet and corrected it.

I used an online disassembler (https://onlinedisassembler.com) and selected Z80 as the processor platform and set the base address of the HEX code to E000 (to match where I suspect the code is located in the memory map). I then copied and pasted the 256 HEX bytes of ASCII code into the window and I got an instant disassembly. I downloaded the resulting disassembly and copied and pasted it into my editor.

I am now going through the code and:

a) Comparing it to my original hand disassembly and
b) Comment the code the best I can.

Dave
 
I was following your other machine with 1702A's and see you found another with similar 1702A's Where are you finding these things.
When I was first trying to read 1702A's, I made a simple board with a dip switch and LEDs to read the data. It takes about 20 minutes to read all 256 locations.
I have a small bench supply for the voltages.
I now can read 1702As with my SIM4 setup. It is still slow as it is in BPNF format over 110 baud. It takes about 5 minutes to read 256 bytes.
Dwight
 
daver2 said:
I have a 'raw' disassembly now. It looks good. I compared it to the hex dump we had previously, and only one (1) byte stood out as different. I tracked this down to a fault in my spreadsheet and corrected it.

I am now going through the code and:

a) Comparing it to my original hand disassembly and
b) Comment the code the best I can.

Dave
Click to expand...

Wow! I didn't expect such rapid progress! Thanks Dave! I was just thrilled to see this board appeared to work, and that I now have a way to dump 1702As. Let me know if you see anything that resembles a prompt.. anything we should be looking for on the Microkit that indicates it's at least trying to run the ROM.

I've still had no luck on software. One fellow reached out but the tape he had had been overwritten. It seems to be a very rare machine today, despite evidence it was produced for a time.
 
Dwight Elvey said:
I was following your other machine with 1702A's and see you found another with similar 1702A's Where are you finding these things.
When I was first trying to read 1702A's, I made a simple board with a dip switch and LEDs to read the data. It takes about 20 minutes to read all 256 locations.
I have a small bench supply for the voltages.
I now can read 1702As with my SIM4 setup. It is still slow as it is in BPNF format over 110 baud. It takes about 5 minutes to read 256 bytes.
Dwight
Click to expand...

I find them usually on ebay. I haven't found another source for things. I've tried to divine how some of the other collectors here get stuff, really rare stuff that seems to just drop into their laps with regularity. But they won't disclose their methods. So yeah, 99% is just me on ebay seeing stuff because I've got a search that alerts everything coming up under vintage computing. Since my interest is older stuff I tend to end up with machines that use the 1702.

I actually acquired a reader/burner made by MSI, but I've not been able to find the software. I'm not skilled enough to build my own. Stumbling on this digital group eprom board was a lucky strike.. it was in amongst some other random gear.
 
falter said:
Wow! I didn't expect such rapid progress! Thanks Dave! I was just thrilled to see this board appeared to work, and that I now have a way to dump 1702As. Let me know if you see anything that resembles a prompt.. anything we should be looking for on the Microkit that indicates it's at least trying to run the ROM.

I've still had no luck on software. One fellow reached out but the tape he had had been overwritten. It seems to be a very rare machine today, despite evidence it was produced for a time.
Click to expand...

OK. I have disassembled and commented the 256-byte 1702A EPROM.

It is starting to make sense.

There is no 'MONITOR' stored within it - just a tape loader. That's purely it.

The VDU is initialised to display memory starting at 0x8000 by the looks of it. I got that wrong initially.

Interrupts are not used - the I/O is purely polled.

I am trying to understand the tape interface from the code. I 'think' it is a 'bit banging' interface - so it reads one bit at a time from the I/O port and appears to assemble the data bytes into memory (but in a block structured manner).

I will have a further look later...

Dave
 
Interesting! So it sounds like it doesn't give you a 'prompt' like my digital group machine does.. nothing to say 'I'm working' or 'Press play on tape'.

Man I hope I find some software one day.
 
Correct - no prompt, no messages no nothing...

Just 'LOAD and GO' (or HALT on error).

I was having my tea and thought of a problem. We have a catch-22 situation. I can reverse engineer the tape format from the bootstrap ROM - but I have no idea how to write a tape (as that code is not contained within the bootstrap ROM). As we don't have a circuit diagram for the serial/tape card - it will be difficult to 'make it up'.

I could create a replacement monitor (to fit in another 1702A EPROM) so that we can dump from memory, load to memory and execute test programs. I think I know enough to be able to 'drive' the keyboard and VDU (so to speak).

How many bytes is the digital group machine's debug monitor out of interest? I am looking for a dead simple monitor to 'borrow' for the time being...

I might have a look at the 8008 monitor I have somewhere and see if I can 'butcher' that...

I assume that EPROM programmer you purchased should be able to program 1702A EPROMS?

I have just uploaded my EPROM disassembly to https://drive.google.com/drive/folders/13uRBisXNUFSV3QakhvfwUmIIz9EUZf7-?usp=sharing for you to peruse at your leisure...

Dave
 
I suppose I could devote a day to reverse engineering the tape circuit, but it might be dodgy given my limitations. I think the boards are just double layer though so it might be possible.

I had thought of borrowing from digital group's 8080 op sys, but of course you still face the problem of loading it in via tape. The 1702A the dg system uses is just a loader like this Microkit's to pull the 'OS' from tape. I don't know the exact size but the listing for it is here.

I think we might be stuck without knowing how that tape interface works. I even reached out to the current company in desperation but of course they have no records further back than 10 or 15 years or so.

As for the programmer.. I have all the necessary hardware, however I lack the software MSI provided to actually use it. It uses the MSI 6800's SS30 parallel board somehow. MSI software sadly is vanishingly rare.
 
Last edited:
I'd be curious to see the guts of the programmer. A few voltage checks would show if it was for 1702As, 2708 or 2716s all three use completely different voltages.
If the code has a write, it is not an impossible job to install a working CP/M2.0. I've done it.
I assume you have RAM starting at 0 and going up until it hits ROM. If so, it is not too hard to deal with.
Loading CP/M from a tape is a drag though. It really needs a floppy drive.
I do wonder why it has so many EPROMs. I'd think a tape loader is only about 512 bytes.
Dwight
 
The programmer is definitely for 1702as.. I have an original catalog for it. But for sure I'll open it up and take some pics. It's quite large for what it is. I also have a homebrew 1702a programmer that I bought with the digital group card but no cables and no idea how it worked.

I didn't think CP/M was available on tape. Hmm. I'm not sure if I'm picking the right starting address to start at either. I should 'dump' it to pictures and if others want to figure out the deal with it that would be cool.
 
falter said:
The programmer is definitely for 1702as.. I have an original catalog for it. But for sure I'll open it up and take some pics. It's quite large for what it is. I also have a homebrew 1702a programmer that I bought with the digital group card but no cables and no idea how it worked.

I didn't think CP/M was available on tape. Hmm. I'm not sure if I'm picking the right starting address to start at either. I should 'dump' it to pictures and if others want to figure out the deal with it that would be cool.
Click to expand...

A machine with full control of the tape, forward, reverse, read and write could have been done on tape. It would have required a lot of software overhead if done on the host computer but a separate processor on a tape controller could have been done. One could use a PC as a disk controller and run CP/M through a serial link, with out too much overhead.
My feeling is that it may have used one of the early disk controllers, like the Digital Systems. These had a controller in the drive box and only needed a bus buffer board in the main machine. The thing I like about these is that there is no software needed in the machine. It would detect a reset signal and DMA the first track in memory. In my IMSAI, I have 64K of RAM and no lost space for ROMs or other firmware.
Dwight
 
You must log in or register to reply here.
Back
Top