Re: CBM-900 floppy disk format/encoding

Hi Simon,

Thanks for your answer

On 2012-01-09 21:43, Simon. Laule wrote:
> Am 09.01.2012 21:11, schrieb Uffe Jakobsen:
>>
> ...
>> The 6508R2 along with the FDC chip 8723 and the ROM found between them
>> constitutes the floppy subsystem - just like every other external CBM
>> drive we know of - only with the difference that the DOS in its ROM is
>> not CBM DOS as we know it from the external drives. The subsystem
>> seems to resemble a traditional simple disk/floppy subsystem much more
>> than any CBM diskdrive that I know of - and yet is has its own CPU for
>> some unknown reason.
> The reason is quite simple:
> Due to the way unix schedules the processes, it cannot be guaranteed
> that the floppy gets enough CPU to read data
> fast enough. (Remember: We are talking about 1983 technology)
>
> Based on the functional description in the datasheet(*), the floppy
> subsystem was meant to be on a separate board,
> connected back to the 8716 DMA on the mainboard. Between unix and the
> subsystem was a SCSI-like protocol
> planned, perhaps in a way to reuse as much of the harddisk connection as
> possible.
>

Well the 8716R1 (PDMAC) Physical DMA Controller is there it is labled 
U79 just left of the Z8001 (U1) on this picture it is to the right on 
the edge of the picture

http://www.datamuseum.dk/foto/20111117/dsc_9785

What you describe about the separate board is in fact the WD-1003 OEM 
licensed MFM harddisk controller that is mounted upside down just below 
the bridge that carries the floppy drive

See pictures:

http://www.datamuseum.dk/foto/20111117/dsc_9835
http://www.datamuseum.dk/foto/20111117/dsc_9836

This harddisk controller has a 34 pin pibbon cable that is connected to 
CN7 just above the floppy onboard circuit but looking at the pcb most of 
its wiring seem to end up in the 8716

http://www.datamuseum.dk/foto/20111117/dsc_9789

BTW: the (two) floppy connectors CN202 and CN203 (one unused - draws a 
fan) only uses a custom 24 ribbon cable

> (*)I will try to copy as much as possible in a file later, unfortunately
> most of the pages are unreadable
> due to fact that it was made on a bad xerox machine long time ago.
>

Thanks that would be much valuable for us.

Kind regards Uffe :-)




       Message was sent through the cbm-hackers mailing list