Page 1 of 1

Motorola VERSADOS 5.25" FD imaging

Posted: Mon Mar 15, 2021 12:41 am
by TonyF
has anyone made a KF script for imaging Motorola Versados 5.25 " disks.
Seem to think i have read that someone has already done this, reported on a KF PR blurb

have been trying and get everything (16 MFM 256 byte sectors) except side 0 first track using MFM.
the ones in green seem to be virgin sectors from original disk formatting and written sectors have artifact data in formatting headers

even using FM on this S0T0 yields nothing except a mismatch error here
Maybe there is interleave here on track 0 to speed up booting.
In this case I'll need some tuition on how to address this in KF script.

is there a user library of KF scripts for recovering images anywhere ?

see two image attachments.

Re: Motorola VERSADOS 5.25" FD imaging

Posted: Thu Mar 18, 2021 7:55 am
by ZrX
TonyF wrote: Mon Mar 15, 2021 12:41 am have been trying and get everything (16 MFM 256 byte sectors) except side 0 first track using MFM.
the ones in green seem to be virgin sectors from original disk formatting and written sectors have artifact data in formatting headers

even using FM on this S0T0 yields nothing except a mismatch error here
Maybe there is interleave here on track 0 to speed up booting.
In this case I'll need some tuition on how to address this in KF script.
Off the top of my head I recall that dumping mixed format media into a single imagefile is not possible. You'd have to dump the first track from side 0 as FM (you may also have to force a sector size for it to decode correctly. Then dump the rest of the disk as MFM and finally you need to manually combine data from the two dumps.

Such mixed format imagefiles may or may not work in emulated environments.

Re: Motorola VERSADOS 5.25" FD imaging

Posted: Thu Mar 18, 2021 7:14 pm
by SomeGuy
Could you post the disk dump?

A mistmatch error would suggest a different sector size. That is typical when encoding differs.

Re: Motorola VERSADOS 5.25" FD imaging

Posted: Mon Mar 22, 2021 2:02 am
by TonyF
Made some progress, need to find a way to merge these in a usable way. I need to get to a versados system and see in what form the file index stores the data data pointer and storage attributes.

this reads the disk data body. all except t0s0 is read properly:
dtc -fVERSAdos_d.img -v300 -z1 -i4 -n+Z16 -e79 -l8
each sector here is 256 byte MFM at 16 sectors per track it zero fills the t0s0 it cannot read (4096 bytes)

this reads the disk ID and Index t0s0:
dtc -fVERSAdos.I.img -v300 -z0 -g0 -i3 -n+Z16 -e0 -l8
This contains 16 128 byte FM sectors only but produces a monster output file, the data is only 2048 bytes

Re: Motorola VERSADOS 5.25" FD imaging

Posted: Mon Mar 22, 2021 7:16 am
by TonyF
All cracked now - thanks for the hints guys

Need to remove the first 4k bytes from the body data image and replace it with the first 2k bytes from the index image. This restores the sector numbers (VDOS works in sector numbers of 256 bytes and most stuff gets plonked on to a sector start address. On the attached image multiply a PSN by 256 to get its byte address in this image.

I made a file using a running version of this disk image that is a multi segment file "POINTERS.SA" its file control block points to a bunch of segments forming the bits of the file. So "pointers" directory entry point to a file control block that has the pointers to all the parts in start to end order.

I tried to attach a Versados "Repair " command manual PDF but it exceeds the 2mb limit on the BB here. This has all the data on Versados file structures It's in a 4.5 mb zip file, if this BB has a ftp hole I'll send it if you give me the URL.

Re: Motorola VERSADOS 5.25" FD imaging

Posted: Mon Mar 22, 2021 7:38 am
by TonyF
Oops:

Sorry it is POINTERS.LS : points to PSN 59B (59B00 in this disk image) for file control block which sprays you around to get the file parts for this big text file.