Close

FFFFIIIINNNNAAAALLLLLLLLYYYYYY

A project log for OMNI 4 - a Kaypro 2x Logic Analyzer

A while back I acquired a rare logic-analyzer, whose lone system-diskette needed backing-up. Now this page is all things OMNI 4

eric-hertzEric Hertz 03/29/2017 at 06:345 Comments

Sheesh

So check-this...

After the last experimentation trying to access the image-file via cpmtools, I was somewhat convinced the image-file is right. Though, I wasn't able to figure out the *exact* specifications of the high-level disk-format (these things the ROM and OS *on the disk* would know, but have to be specified by the user if using a different system), I was able to extract some files via 'cpmcp' and they seemed mostly-alright. Again, the specific settings I don't know (skew, block-size, etc) are things actual system would know... so if I have the image written to disk properly I don't need that information to use the disk in the system it's designed for, only if I want to use that disk-image on *another* system.

So, since I was able to [mostly] browse files (and a few other things including hexdumps of the image-file), I figured the disk-image *must* be right. Right?

So why didn't it work? Last attempt was using gap-sizes that worked with a different disk-image, downloaded off the 'net, so it shoulda worked, right?

So, now I'm writing a whole bunch of floppies with the image I made... using different gap-sizes... Then I'll plop the drive back in the system and try 'em out until one works...

But then I notice something weird...

WTF... it's writing sectors 10-19 on Head 0 (and 0-9 on Head 1)!

ImageDisk has a utility 'bin2imd' which takes a raw binary file containing all the sectors of a disk and allows you to convert that to the IMD format... What I did with the raw sectors, of course, was to 'cat [eachSector] >> image.bin' (in order) then used 'bin2imd' to create a floppy-disk image.

You can imagine, bin2imd needs some options, to indicate the sector-header information, etc.

So... turns out the sector-map options are *reversed*. If you say "sm0=0-9 sm1=10-19" you'd think the sectors on head 0 would be numbered 0-9 and those on head 1 would be 10-19, right? Wrong.

...

And now it works. And I can finally reclaim my couch and maybe even my coffee table!

-----

The sectors aren't interleaved... it's noticeably slower than with the original disk. We'll see if I have the patience to change that... but that's just a matter of options with ImageDisk (and swapping the drive between machines, which I've tired of, or finally figuring out how to use a 1.2MB drive).

Discussions

Thomas wrote 03/29/2017 at 18:18 point

Congrats! Now, finally, you can have your life back :P

  Are you sure? yes | no

Eric Hertz wrote 03/30/2017 at 05:39 point

LOL! Not so fast, there's a new unexpectedly-ridiculously-difficult venture in the next log!

  Are you sure? yes | no

Thomas wrote 03/31/2017 at 18:03 point

OK, I guess that the cat doesn't mind. The space between the sofa and the board looks just right.

  Are you sure? yes | no

Eric Hertz wrote 04/01/2017 at 04:59 point

haha, now that you mention it, I'm surprised I haven't seen her under there! Must be all the doodads taking up space, I'll clean that up first :)

  Are you sure? yes | no

Yann Guidon / YGDES wrote 03/29/2017 at 09:39 point

Congratulations !

You're now a floppy disk certified specialist :-P

  Are you sure? yes | no