Date   : Sat, 06 Jan 2007 01:47:38 -0000
From   : BBCMailingList@... (Ian Wolstenholme)
Subject: Superform Problem

I've cracked it, I'm sure you will all be delighted
to know that I have just managed to format this
Acorn Winchester Disc successfully.  Superform
took 7 re-formats to do it, finding 52 bad sectors
in all and at last I've got a 30MB disc formatted
as a 30MB disc!

I had to play around with Superform a bit.  I
wasn't too wide of the mark when I said I
thought it was a problem with the defect list.
In fact, it was down to the interleave.  I didn't
realise that Superform always uses an interleave
of 5, there is no way for the user to input a
different value without editing Superform.

For some reason the Rodime disc didn't seem
to like an interleave of 5, somewhere along
the line it got confused when translating the
defect list into physical locations and either gave
me a "bad argument" error when trying to
re-format or a verify error &90 at every single
sector from a certain point onwards.

In the end I used an interleave of 2 which did
the trick, although I had to amend the calculation
which works out the physical location of a bad
sector when building the defect list to take into
account the different interleave.  I think there
may even be an error in the original translate
calculation in Superform as the figure used doesn't
seem to be the figure you would expect on an
interleave of 5.  

Now I can try restoring one of those Digistore
tapes to the Winchester...

Best wishes,


Ian
 
----- Original Message -----
From: Ian Wolstenholme [mailto:BBCMailingList@...]
To: bbc-micro@...
Sent: Thu, 4 Jan 2007 23:13:29 -0000
Subject: [BBC-Micro] Superform Problem

I've got out the Acorn Winchester Disc again
to see if I can get it formatted correctly.  When it
arrived, it was formatted as a 10MB drive but in
reality it's a 30MB disc.  However, try as I might,
Superform won't format it with the correct disc
parameters.

Usually what happens is that the first format is
OK then the verify process shows up about 35
bad sectors, so it reformats.

Something seems to go wrong on the re-format
because I can hear the disc "ticking" as if it is
working but then it seems to re-calibrate and
starts "ticking" again and repeats this process
about ten times before Superform stops, giving
Error 24 in Format, which is Bad Argument
according to the Adaptec error list.

I've tried various things to get round this without
any success.  It seems to be a problem with the
defect list because if I manually reduce the size
of the defect list by altering the variable pointing 
to the length of the defect list and then format
again, or if I start again without a defect list,
the first format works and then it falls down
on the re-format after verifying.

I've had this problem before with other old
Winchesters but I have usually got them to
format successfully eventually either by
perservering, or I have changed the drive
attached to the controller and then gone back
to the original drive which seems to have the
effect of clearing the fault.

In the case of this Acorn Winchester, however,
I must have formatted it about 20 or 30 times
in total before Christmas and today with the
same error happening every time.  Trying the
disc with a different controller or vice-versa isn't
really an option given my highly dubious track
record of breaking things by taking them to
bits.

The only way I can make the disc "useable" is
to call the routine which writes the free space
map and the root directory after the format
breaks down but presumably this is leaving 
the disc in a less than satisfactory state in
terms of known defects not being mapped out
at the format stage.

Has anyone else come across this kind of
problem with Superform and is there a
solution?

Best wishes,



Ian