Page 1 of 3

Request: Hard-sectored disks (5.25) for test material to KF team

Posted: Sat Dec 09, 2017 8:28 pm
by Rakki
Kryoflux team is looking for hard sector floppies (5.25") for test material to improve Kryoflux software support.

So far we have heard about 5, 6, 7, 8 and 13 index holes hard sector disks. All kinds of hard sectored disks are welcome but most importantly they should be good working order and run on real system. Disks must have more than one index hole - otherwise it is soft sectored. See the link below so you can see the difference.

Soft vs hard-sectored floppy:
http://www.minuszerodegrees.net/diskett ... ctored.jpg

Northstar and Heath were commonly known to use hard-sectored floppies but is there other systems? Apple II used sometimes hard-sectored floppies because its disk drive doesn't have the sensor for index hole at all. Especially if there's PC based systems that actually used hard-sectored disks are very interesting.

If you have got test material please reply to this thread. For more details like shipping addess we can use forum's private messages.

Based on Stack Exchange answer:

Hard sectored 8" and 5-1/4" floppy disks were used in some early computers including:
  • DEC Pro 350
  • Heath/Zenith H-89a
  • MITS Altair 8800BT
  • NorthStar Horizon
  • Vector Graphic computers
  • Processor Technology add-on for the Sol-20 computer called the Helios II Disk Memory System
  • Compudata Exidy Sorcerer add-on floppy drive based on the Micropolis
  • Quite some of the Zilog dev systems (MCZ 1/15, MCZ 1/30, MCZ 1/05)

Re: Request: Hard-sectored disks (5.25) for test material to KF team

Posted: Mon Dec 11, 2017 10:39 pm
by IFW
We are looking to at least support reading data written to hard-sectored disks for systems that actually did not use the index signal at all - ie many 8-bit systems.
Apparently, it was at one time popular to write onto these disks, simply because they became dirt cheap after the hard-sectored format was pretty much abandoned in favour of the soft-sectored formats, and it did not matter for drives which did not use index signals anyway.
We might be also looking into supporting "real" hard-sectored formats (where the sector index information was actually used), but reliable/low-level information is pretty scarce on those, and we simply do not have any such disk.

Re: Request: Hard-sectored disks (5.25) for test material to KF team

Posted: Sat May 11, 2019 2:41 pm
by aquarius
I have a System from 1982 named Questar/M from Honeywell Bull that uses hard-sectored 5,25" Floppy disks.

With the Kryoflux of a friend of mine we were able to read the raw data. I would like to give this data to you for analysing as with MFM and FM-Reading mode Kryflux did not give us any image file out.

Post me a PM so I can give you this data.

Best regards
Aquarius

Re: Request: Hard-sectored disks (5.25) for test material to KF team

Posted: Tue May 14, 2019 7:14 pm
by ZrX
After having looked at two Questar disk dumps the format seems as follows:

- 2 sides
- 80 tracks per side
- 16 sectors per track (0-15 on side 0, 16-31 on side 1)
- 256 bytes per sector
- 300 rpm
- MFM sync ...AAAA AAAA AA91 2AAA (bytes 00 00 05 00)
- Decoded track format 05 00 TT SS + data*256 + two checksum bytes + padding
- First checksum byte is TT + SS + data + overflows of the calculation
- Second checksum byte calculation not solved yet

Re: Request: Hard-sectored disks (5.25) for test material to KF team

Posted: Wed May 15, 2019 6:20 pm
by aquarius
:D Great Stuff!

Thank you very much!

Re: Request: Hard-sectored disks (5.25) for test material to KF team

Posted: Thu May 16, 2019 10:41 am
by brightcaster
Wasn't the Questar deliverd with CPM 2.2? I just checked that there isn't any "questar" entry in the diskdefs of cpmtools, but there are some with the parameters mentioned by ZrX. So cpmtools could help to get the files of the disks (if that is want you want)...

David

Re: Request: Hard-sectored disks (5.25) for test material to KF team

Posted: Thu May 16, 2019 5:39 pm
by aquarius
That is a good hint. Didn't know CPM Tools.
At the moment I am happy to have Images of the boot disks thanks to @ZrX.

The next step is to communicate via the serial Interface.

Yes it is CP/M 2.23 and Prologue (QM OS).

Here are the startscreens: (Post #10)
https://forum.classic-computing.de/foru ... well-bull/

Aquarius

Re: Request: Hard-sectored disks (5.25) for test material to KF team

Posted: Thu May 16, 2019 7:39 pm
by ZrX
That looks like some ancient hardware.

The system ROMs might reveal how the full disk checksumming works, if there were dumps of the ROMs and knowledge of the CPU architecture.

Re: Request: Hard-sectored disks (5.25) for test material to KF team

Posted: Thu May 16, 2019 9:02 pm
by brightcaster
The CPU is Zilog Z80 and the 2708 eprom should be readable in 2716 mode while disconnecting pin 19 which is A10 on the 2716 but VDD on the 2708! You get the 2708 data twice than of course in the lower and upper half of the 2716 file...

EDIT! Sorry but I was wrong the original Intel 2708 needs the -5V on Pin 21 and +12V on Pin 19 even while reading! So no easy adapter!!!



David

Re: Request: Hard-sectored disks (5.25) for test material to KF team

Posted: Fri May 17, 2019 9:08 am
by aquarius
Yes, it is a Z80 system. I can upload the System ROM which is a 2716.

The 2708 sits on the Keyboard pcb - I think this is the character ROM.