<< Previous Message Main Index Next Message >>
<< Previous Message in Thread This Month Next Message in Thread >>
Date   : Tue, 06 Jul 2004 09:49:39 +0000
From   : Jules Richardson <julesrichardsonuk@...>
Subject: Re: Acorn Winchester unit

On Tue, 2004-07-06 at 08:53, andrew.chesterton@... wrote:
> If anybody is local to the Midlands in the UK and wants to check 
> the BBC - SCSI accesses, I have an Ancot SCSI bus analyser that 
> I am prepared to lend out.

Interesting. Don't ever throw it out!

> I don't have the BBC hardware or the time at the moment to do it 
> myself, but it is something that I have been thinking about for a 
> while now.
> 
> One possible project that I have been thinking about would be 
> using an Altera FPGA to replace the Winchester interface board. 
> This would be able to sit between the 1MHZ bus and the SCSI 
> device(s). It may also be powerful enough to do some address 
> translation and cope with the 256/512 bytes per sector problem 
> and alow 2 BBC writes to 1 SCSI disk sector.

To be honest I'm still thinking that a software solution is better
though - as proved, the Winchester board isn't that complex and can be
built at home with a bit of time. No custom chips and no programming. 

ADFS code has been pretty comprehensively reverse-engineered. The only
downside to a software approch is that you do lose half the disk, but
then when you'd have trouble filling 40MB up with a BBC and an 80MB
drive is about the smallest obtainable these days, that isn't the end of
the world. 

Possibly someone could modify ADFS to work with 512 byte sectors though,
and just chuck away half of the 512 bytes on a read. Of course it does
imply that you lose half your transfer rate on reading though as you're
transferring 256 bytes that you don't need. 

> Can somebody invent a 36 hour day please, then I might get time to do 
> things again....

work in octal and you're close :-)

cheers

Jules
<< Previous Message Main Index Next Message >>
<< Previous Message in Thread This Month Next Message in Thread >>