Dimension 68000 CP/M-86 diskdef

Steven Hirsch

unread,

Sep 2, 2022, 9:33:14 AM9/2/22

to

In case anyone runs across one of these oddball machines:

diskdef dim68k-cpm86
seclen 512
tracks 40
sectrk 20
blocksize 2048
maxdir 64
skew 1
boottrk 1
end

Bill Shen

unread,

Sep 3, 2022, 10:26:32 PM9/3/22

to

Should the topic title be:
"Dimension 68000 CP/M-68K diskdef"?

Steven Hirsch

unread,

Sep 5, 2022, 9:14:32 AM9/5/22

to

On 9/3/22 22:26, Bill Shen wrote:
> Should the topic title be:
> "Dimension 68000 CP/M-68K diskdef"?

They use the same diskette layout for generic (non-Kaypro) CP/M-80, CP/M-86
and CP/M-68k. But, yes, it might have flowed better that way :-).

Martin

unread,

Sep 5, 2022, 11:48:32 AM9/5/22

to

I also analysed the format, while repairing
the corrupted trisoft archive.

I used the diskdefs to extract the FORTRAN files
from the imaged in Dunfield's collection "dimen68k.zip".

They were RELOCated binaries, I used them to verify
the correctness of my work.

But beware, depending on the CP/M Version,
they used two different formats!

Basically your diskdef is equivalent to the
Version 1.2 definition.

########################################
# Microcraft Dimension 68000
# Version 1.0x: dim68-10
# Version 1.2x: dim68-12

diskdef dim68-10
seclen 512
tracks 80
sectrk 10
blocksize 2048
maxdir 64
skew 0
boottrk 2
# this format wastes half of the directory entry
logicalextents 1
os 2.2
end

diskdef dim68-12
seclen 512
tracks 80
sectrk 10
blocksize 2048
maxdir 64
skew 0
boottrk 2
os 2.2
end

Steven Hirsch

unread,

Sep 6, 2022, 9:37:56 AM9/6/22

to

On 9/5/22 11:48, Martin wrote:

> I also analysed the format, while repairing
> the corrupted trisoft archive.
>
> I used the diskdefs to extract the FORTRAN files
> from the imaged in Dunfield's collection "dimen68k.zip".
>
> They were RELOCated binaries, I used them to verify
> the correctness of my work.
>
>
> But beware, depending on the CP/M Version,
> they used two different formats!
>
> Basically your diskdef is equivalent to the
> Version 1.2 definition.
>

Thanks, Martin. The version 1.0x definition is essentially the same as Kaypro
10 is it not? That would make sense since Kaypro is one of their standard
emulations for the Z80 board.

Martin

unread,

Sep 6, 2022, 1:04:40 PM9/6/22

to

Am 09/06/2022 03:37 PM, Steven Hirsch schrieb:
> On 9/5/22 11:48, Martin wrote:
>
>
>> I also analysed the format, while repairing
>> the corrupted trisoft archive.
>>
>> I used the diskdefs to extract the FORTRAN files
>> from the imaged in Dunfield's collection "dimen68k.zip".
>>
>> They were RELOCated binaries, I used them to verify
>> the correctness of my work.
>>
>>
>> But beware, depending on the CP/M Version,
>> they used two different formats!
>>
>> Basically your diskdef is equivalent to the
>> Version 1.2 definition.
>>
>
> Thanks, Martin. The version 1.0x definition is essentially the same as
> Kaypro 10 is it not? That would make sense since Kaypro is one of their
> standard emulations for the Z80 board.
>

Did not notice this coincidence, good to know.

I just couldn't read some of the images from bitsavers,
"EMUL_1.04.IMD" is an example.

Every file with more than one extent had only zeroes in it beginning
with offset 4000h, then continues at offset 8000h, and so on.

Have fun!
Martin