Hi James!
On Tue, 9 Oct 2012 10:42:02 -0700 (PDT), jmk wrote:
> This is definitely old. It only has Exit, Return, Save and Poke options,
> no Printing, no Jumping, no Toolkit.
Jup. As said: The woes of an early adopter. Basically, I thought about
designing my own "Multiface", long before the MF-1 appeared on the market.
But those days, you couldn't simply start up EAGLE, load a few libraries,
connect some buses and signals, click on "Autoroute" and send the
CAD-/CAM-Files to some service provider.
And the parts were pretty expensive - I remember having paid big bucks for
the inevitable Z80-PIO and the bigger RAM chips to replace the upper 32k
of my Speccy and allow bank switching. Not to speak of the female part
matching the edge connector...
So erasing/burning an E-PROM, designing a PCB AND developing M/C software
to deal with backupping what the Speccy had in its RAM was way out of
reach. Today, you could emulate stuff instead of crashing GENS for the
1000th time - that's how I developed mdv2img, the tool that helped me to
create images from Microdrive cartridges.
Some 25 years ago, not even a singlestepping mode with register display
existed, so it was pretty much trial and error.
> I tried it with Genie and Lifeguard and, sure enough, it can't start them
> when you press the button.
*LOL* Yes, I failed miserably, too... ;-)
> The picture looks like 1 EPROM, 1 RAM chip (HM6264 = 8K) and possibly a
> load of small logic chips. It's possible that you have the full 8K RAM,
> although I can't tell for sure if all the address lines are attached.
I never disassembled the MF-1 ROM nor did I care much about how it worked.
As long as it worked. Did it copy part of the Spectrum's RAM to it's own
RAM to be able to create the BASIC loader plus SCREEN$ and 48K RAM? I
remember my MF-1 corrupting parts of the screen, so it probably couldn't
create a perfect copy of the snapshot.
> I'll have to upgrade the dumping tool to check the size of the RAM.
Are there any schematics of the MF-1 around? IF-1/2 and Microdrives are
pretty well documented and lately even the ULA has been reverse engineered
(hats off to Chris Smith, he did an unbelievably cool job - see
http://www.zxdesign.info/book/insideULA.shtml for details!) - but I wasn't
able to get info on how the MF-1 worked.
Probably just generating an NMI, paging out the Speccy's ROM and paging in
the MF-1 ROM and RAM instead. What I didn't know is, that you could trigger
that paging mechanism via software (addresses 9Fh/1Fh). And I never
bothered to investigate how an image is actually restored.
> I think you might be able to burn an EPROM of a newer MF1 version,
> install it and get it to work, as long as the full 8K RAM is wired up.
Indeed. But I still have no E-PROM burner. In the age of GBs of Flash RAM,
this probably will have to be stolen out of a museum... ;-)
> ROMS.BIN has the MF128 ROM at offset 0x1E000 (the info is available in
> the DOS version's documents), but it uses different ports from the MF1,
> so you can't swap them.
Ah. OK.
> The CRC is calculated by a routine in contended memory, so it could be
> made to move a little quicker. ;)
Hey, 2s (IIRC) is good enough for me.
> Regarding digital IDs, when I use Outlook Express on my XP machine, it
> gives me the option to get a digital ID and then goes to a dead link!
Lame. I gave up on Outlook years ago. TheBat (Ritlabs) is so much better,
that I never felt the urge to look back.
Cheers,
Volker
P.S.: Make sure to read Andrew Pennel's blog about how he wrote his forst
book, "Master Your ZX Microdrive":
http://blogs.msdn.com/b/andypennell/archive/2009/04/27/you-know-you-re-old-when-the-copyright-on-your-first-book-expires.aspx
Incredible.