New Hard Drive

111 views
Skip to first unread message

Tim

unread,
May 20, 2006, 5:43:03 AM5/20/06
to
Morning All

Having listened to Paul Middleton at Wakefield I quickly bought a
replacement hard drive for my Risc PC.

Can anyone point me to some simple non-tech instructions for effecting the
replacement?

I am assuming that this will be covered somewhere out there on the web but I
can't find it on/via Google.

Thansk in advance

--
Tim

Jess Hampshire

unread,
May 20, 2006, 6:00:24 AM5/20/06
to
In message <75bbea2...@lynndene.demon.co.uk>
Tim <T...@lynndene.demon.co.uk> wrote:

> Having listened to Paul Middleton at Wakefield I quickly bought a
> replacement hard drive for my Risc PC.
>
> Can anyone point me to some simple non-tech instructions for effecting the
> replacement?
>
> I am assuming that this will be covered somewhere out there on the web but I
> can't find it on/via Google.

Are you upgrading or replacing a failed drive?

--
Jess Iyonix
contact http://jess.itworkshop-nexus.net
Hotmail is my spam trap - don't email
valid - mailto:nos...@jess.itworkshop-nexus.net

Paul Bunyan

unread,
May 20, 2006, 7:14:48 AM5/20/06
to

Hi Tim,

I recently danced this merry dance. It is a 2 stage process. The 1st stage
is the dismantling of your RiscPC to remove the old drive and fit the new
drive. The 2nd stage, done with software, is preparing the drive for use.

I took photos when I replced my hard disk. If time permits this weekend I'll
throw together a web-page with step-by-step details.

Regards

Paul B.


Dave Plowman (News)

unread,
May 20, 2006, 8:45:10 AM5/20/06
to
In article <75bbea2...@lynndene.demon.co.uk>,

Tim <T...@lynndene.demon.co.uk> wrote:
> Having listened to Paul Middleton at Wakefield I quickly bought a
> replacement hard drive for my Risc PC.

> Can anyone point me to some simple non-tech instructions for effecting
> the replacement?

Assuming you are on the IDE bus and have at the moment one HD and one CD.
the best thing to do is to temporarily disconnect the CD and replace it
with the new HD - the connections are the same. The HD may just be laid
somewhere safe - it doesn't need to be fixed in place. But use some card
etc to insulate it. Make sure to set the HD to master or slave to match the
CD. Then alter the number of HDs in Boot to 2.

Everything being ok, you should then be able to copy everything across.
Then physically swop the discs and put re-connect the CD.

I'm sure you'll find details of the mechanics of stripping an RPC - it's
pretty simple, but take notes of connectors, etc and label if you're not
sure. Or ask here giving details of the hardware fitted.

One other thing - if you've given your HD a name other than HardDiscD4,
remember to re-name the new one the same.

--
*A plateau is a high form of flattery*

Dave Plowman da...@davenoise.co.uk London SW
To e-mail, change noise into sound.

Tim

unread,
May 20, 2006, 11:35:14 AM5/20/06
to
In message <4e29fb6...@davenoise.co.uk>

"Dave Plowman (News)" <da...@davenoise.co.uk> wrote:

> In article <75bbea2...@lynndene.demon.co.uk>,
> Tim <T...@lynndene.demon.co.uk> wrote:
> > Having listened to Paul Middleton at Wakefield I quickly bought a
> > replacement hard drive for my Risc PC.
>
> > Can anyone point me to some simple non-tech instructions for effecting
> > the replacement?
>
> Assuming you are on the IDE bus and have at the moment one HD and one CD.
> the best thing to do is to temporarily disconnect the CD and replace it
> with the new HD - the connections are the same. The HD may just be laid
> somewhere safe - it doesn't need to be fixed in place. But use some card
> etc to insulate it. Make sure to set the HD to master or slave to match the
> CD.

Can you explain this bit a little further please... How? Where?

Otherwise thats very helpful

--
Tim

Tim

unread,
May 20, 2006, 11:35:57 AM5/20/06
to
In message <1d52ec2...@itworkshop.invalid>
Jess Hampshire <phant...@hotmail.com> wrote:

> In message <75bbea2...@lynndene.demon.co.uk>
> Tim <T...@lynndene.demon.co.uk> wrote:
>
> > Having listened to Paul Middleton at Wakefield I quickly bought a
> > replacement hard drive for my Risc PC.
> >
> > Can anyone point me to some simple non-tech instructions for effecting the
> > replacement?
> >
> > I am assuming that this will be covered somewhere out there on the web but I
> > can't find it on/via Google.
>
> Are you upgrading or replacing a failed drive?
>

Upgrading before disaster strikes...! (I hope)

--

Dave Plowman (News)

unread,
May 20, 2006, 12:09:56 PM5/20/06
to
In article <abf90a2...@lynndene.demon.co.uk>,

Tim <T...@lynndene.demon.co.uk> wrote:
> > Assuming you are on the IDE bus and have at the moment one HD and one
> > CD. the best thing to do is to temporarily disconnect the CD and
> > replace it with the new HD - the connections are the same. The HD may
> > just be laid somewhere safe - it doesn't need to be fixed in place.
> > But use some card etc to insulate it. Make sure to set the HD to
> > master or slave to match the CD.

> Can you explain this bit a little further please... How? Where?

Ah. So you've never had a look inside an RPC?

If you take the top off the RPC, you should see the back of the CDR. There
are three plugs, looking at the back from left to right, a small black one
which is the audio and you can ignore, the large ribbon middle one which
is the IDE cable, and the right hand end semi opaque one which is the
power cable. The jumper for slave or master is marked on the CD so you'll
need to read that - probably slave - but you might have to refer to the
instructions for the HD. The jumper is usually a small female 'plug' -
about 1/4" wide.

--
*Acupuncture is a jab well done*

Tim

unread,
May 20, 2006, 1:01:30 PM5/20/06
to
In message <4e2a0e2...@davenoise.co.uk>

"Dave Plowman (News)" <da...@davenoise.co.uk> wrote:

Well I think I have peered inside maybe once. I am, however, one of what I
expect is a small breed viz a user who has no real interest in the tech side
but just likes the way the whole GUI etc works.

I will have a delve later and report back... if I can!

Thanks for the help

--
Tim

Message has been deleted

Dave Symes

unread,
May 20, 2006, 5:09:34 PM5/20/06
to
Opportune thread.

Can I also ask for some advice on this subject please?

I've never had any trouble adding or replacing Hardrives on this SARPC,
but SWMBO presented me with a Hardrive a couple of days ago and asked if I
would install it in her two slice SARPC as drive 5 for backups.

She only had one drive in the machine so did have a spare IDE channel.

I've done all the usual stuff, set it to slave, and cabled it up.

Reset !Boot to recognise two hardrives, and it does.

Now the problem:

It's a (New) Hitachi DeskStar 40Gig hardrive.

Unfortunately Hform 2.56 will not finish the Formatting process.

The process:

IDE ADFS:5 identifies itself as:
Description: .D.7.8.4.P.A.2
Firmware version: .F.0.2.B
Configuration: 255 Cylinders, 16 Heads, 63 Sectors/Track

This drive does not have a valid ADFS Format. (obviously)


Option (9) for "Other" drive.
255 Cylinders 16 Heads 63 Sectors
Drive parameter init flag 1
Parking Cylinder 245
Option (A) No more changes to the defect list.
Disc will be formatted 254, 16, 63
Parking cylinder 254, Init flag 1, LBA flag 0
Format or Initialise Option (F)
Soak test (N)
Bootable disc (N)
Long filenames (Y) ***She's on RO 4.02***
Are you sure... ADFS:5 (Y)

Formatting ticks over to 254
Large File allocation I've tried both 256 and 512 which are at this point
accepted.

Writing Defect list
Creating Map
Writing Root directory
HForm failed
Disc not understood has it been formatted?
At line 71

Press return etc.


I've tried to just initialise, and reformat twice, all failures.

Any useful suggestions, please?

Her machine is running RISC OS 4.02
The version of HForm I'm using is 2.56

Thank on her behalf.
Dave S

--

druck

unread,
May 20, 2006, 6:08:49 PM5/20/06
to
On 20 May 2006 Dave Symes <d...@ukgateway.net> wrote:
> Unfortunately Hform 2.56 will not finish the Formatting process.

Somewhere in the process you need to select LBA mode, otherwise the shape
will be wrong. HFROM *should* still format it, but it be the wrong size
(probably 8GB).

---druck

--
The ARM Club Free Software - http://www.armclub.org.uk/free/
The 32bit Conversions Page - http://www.quantumsoft.co.uk/druck/

Andrew Hill

unread,
May 20, 2006, 6:40:25 PM5/20/06
to
Dave Symes wrote:
> Opportune thread.
>
> Can I also ask for some advice on this subject please?
>
> I've never had any trouble adding or replacing Hardrives on this SARPC,
> but SWMBO presented me with a Hardrive a couple of days ago and asked if I
> would install it in her two slice SARPC as drive 5 for backups.
>
> She only had one drive in the machine so did have a spare IDE channel.
>
> I've done all the usual stuff, set it to slave, and cabled it up.
>
> Reset !Boot to recognise two hardrives, and it does.
>
> Now the problem:
>
> It's a (New) Hitachi DeskStar 40Gig hardrive.
>

Yes. I discovered previous problems with a Hitachi 60Gb deskstar which I
investigated. Sorry the below stuff is a bit techy; you should be able
to extract information to the depth you're interested in though!


I had a long chat with the Hitachi (well, IBM as-was) firmware team
about the problem, who were extremely friendly.

To demonstrate the age of our equipment, the ADFS bus is using a legacy
feature from before the ATA-1 specification (that spec was written
1990ish IIRC, don't quote me; pretty much BC in hard disc terms anyway,
and significantly pre-dating the RISC PC!) The feature determines
whether to perform 16-bit, or 32-bit transfers. If the RISC PC can't
detect the difference, it uses 16-bit transfers.

Oddly enough, that was deprecated 16 years ago, but we still use the
feature, and we're reading a now 'unallocated' pin, which may be high or
low. As a result, it fluctuates, but is usually set to mean 16-bit
transfers.

Why is this techy stuff important? Well, if you try and format it on
HForm, it will read the drive size as a 16-bit value, and then gets the
size wrong and that's as far as you get. Yay.

However, if it correctly does identify the drive in 32-bit transfer
mode, or worse still you move a drive across which has been
pre-formatted, you can get a drive which works intermittently, and then
causes data corruption when it flips back to 16-bit mode mid-transfer. Erk.


Newer interfaces (ie the third-party ones), don't do this - they will
happily correctly work with the 60Gb Deskstar drive I've got. What I
cannot tell is whether this is a hardware issue (because of the
motherboard chipset,) or whether it's due to behaviour of ADFS itself.

From what I've seen, fairly recent (~20Gb) Maxtors and Seagates appear
to work correctly on the ADFS bus. However, the sensible alternative
would of course be a Unipod, which adds lots of other things as well and
will give you noticeably better hard disc access performance ;-).

Best wishes,

Drew

David Holden

unread,
May 21, 2006, 1:27:21 AM5/21/06
to

On 20-May-2006, Dave Symes <d...@ukgateway.net> wrote:

> Opportune thread.
>
> Can I also ask for some advice on this subject please?
>
> I've never had any trouble adding or replacing Hardrives on this SARPC,
> but SWMBO presented me with a Hardrive a couple of days ago and asked if I
> would install it in her two slice SARPC as drive 5 for backups.
>
> She only had one drive in the machine so did have a spare IDE channel.
>
> I've done all the usual stuff, set it to slave, and cabled it up.
>
> Reset !Boot to recognise two hardrives, and it does.
>
> Now the problem:
>
> It's a (New) Hitachi DeskStar 40Gig hardrive.

That's almost certainly the problem. Many of these drives won't work
correctly when connected to a RiscPC motherboard.

I'm afraid you've just discovered (too late) the advantages of buying from
someone who understands the idiosyncrasies of Acorn computers :-(

--
David Holden - APDL - <http://www.apdl.co.uk>

Dave Symes

unread,
May 21, 2006, 1:00:50 AM5/21/06
to
In article <38022f2a...@druck.freeuk.net>,

druck <ne...@druck.freeuk.com> wrote:
> On 20 May 2006 Dave Symes <d...@ukgateway.net> wrote:
> > Unfortunately Hform 2.56 will not finish the Formatting process.

> Somewhere in the process you need to select LBA mode, otherwise the shape
> will be wrong. HFROM *should* still format it, but it be the wrong size
> (probably 8GB).

> ---druck

Thanks for the info Mr D.
I guess what you are really saying is, she's bought a bum HD type as far
as RO is concerned.

Kinda weird considering I have 40Gig Seagate as Drive 4 on this machine.

Anyway again thanks
Dave S

--

Dave Symes

unread,
May 21, 2006, 3:52:42 AM5/21/06
to
In article <446f9ad8$0$2654$ed26...@ptn-nntp-reader01.plus.net>,

Andrew Hill <us...@example.net> wrote:
> Dave Symes wrote:
> > Opportune thread.
> >
> > Can I also ask for some advice on this subject please?
> > I've never had any trouble adding or replacing Hardrives on this
> > SARPC, but SWMBO presented me with a Hardrive a couple of days ago and
> > asked if I would install it in her two slice SARPC as drive 5 for
> > backups.
> >
> > She only had one drive in the machine so did have a spare IDE channel.
> >
> > I've done all the usual stuff, set it to slave, and cabled it up.
> >
> > Reset !Boot to recognise two hardrives, and it does.
> >
> > Now the problem:
> >
> > It's a (New) Hitachi DeskStar 40Gig hardrive.
> >

> Yes. I discovered previous problems with a Hitachi 60Gb deskstar which I
> investigated. Sorry the below stuff is a bit techy; you should be able
> to extract information to the depth you're interested in though!

[Snip very useful info]


> Newer interfaces (ie the third-party ones), don't do this - they will
> happily correctly work with the 60Gb Deskstar drive I've got. What I
> cannot tell is whether this is a hardware issue (because of the
> motherboard chipset,) or whether it's due to behaviour of ADFS itself.

> From what I've seen, fairly recent (~20Gb) Maxtors and Seagates appear
> to work correctly on the ADFS bus. However, the sensible alternative
> would of course be a Unipod, which adds lots of other things as well and
> will give you noticeably better hard disc access performance ;-).

> Best wishes,
> Drew

Andrew,
Thanks for that run down, I did understand what you've written.

Interesting, considering I have a 40 Gig Seagate as my Drive 4 on this
machine. (Also SARPC but running Select 4.39 instead of 4.02) connected
to the ADFS bus and formatted last year with HForm 2.56 not a problem.

The bottom line is, at least on her machine, type wise, she has a Bum
Steer.

She does have a Unipod on the machine, with USB and Networking active... I
suppose I'd better get the IDE update for her and use that as the
interface.

Thanks again
Dave S

PS: Brings forth another question, what do you use to format a HD
connected to the Unipod IDE interface?
D.

--

Dave Symes

unread,
May 21, 2006, 4:05:31 AM5/21/06
to
In article <4dac1tF...@individual.net>,
David Holden <black...@apdl.co.uk> wrote:

> On 20-May-2006, Dave Symes <d...@ukgateway.net> wrote:

> > Opportune thread.

[Snip]

> >
> > It's a (New) Hitachi DeskStar 40Gig hardrive.

> That's almost certainly the problem. Many of these drives won't work
> correctly when connected to a RiscPC motherboard.

> I'm afraid you've just discovered (too late) the advantages of buying
> from someone who understands the idiosyncrasies of Acorn computers :-(

There's the rub...

I bought my 40 Gig Seagate from the Tardis shop (CJE) which is probably
why it formatted and is still okay.

Hers came from one of those big barn places (PCW) just down the road.

Anyway, taking Andrew's advice, All is not lost, I'll get her to phone the
Tardis Shop on Monday and get an IDE upgrade for her Unipod.

Thanks for the input.

Cheers
Dave S

--

Theo Markettos

unread,
May 21, 2006, 5:59:18 AM5/21/06
to
Andrew Hill <us...@example.net> wrote:
> I had a long chat with the Hitachi (well, IBM as-was) firmware team
> about the problem, who were extremely friendly.
>
> To demonstrate the age of our equipment, the ADFS bus is using a legacy
> feature from before the ATA-1 specification (that spec was written
> 1990ish IIRC, don't quote me; pretty much BC in hard disc terms anyway,
> and significantly pre-dating the RISC PC!) The feature determines
> whether to perform 16-bit, or 32-bit transfers. If the RISC PC can't
> detect the difference, it uses 16-bit transfers.

Hmm... interesting. I'm not disbelieving you, but that seems a little odd
because ATA-1 and standards until much later (mid-90s) didn't support 32 bit
transfers. Are you sure it wasn't 8 bit transfers? There's an IOCS16 pin
which is used to denote 16 bit transfers: some early drives (20MB, 40MB)
could be talked to in 8 bit mode which was designed for PC XTs and similar.
IDE was designed for 16 bit machines so no-one considered a 32 bit
implementation until sometime later.

ATA-1 is here:
http://t13.org/project/d0791r4c-ATA-1.pdf

/IOCS16 is an output from the drive, pin 32 on the IDE bus, which is
connected to /IOCS16 input (pin 27) on the FDC37C665, and pulled high with
R46, a 1K2 resistor. It's used to generate strobes to latch either 8 bit or
16 bit values from the drive onto the Risc PC's IO databus.

In the draft for ATA-8:
http://t13.org/docs2005/d1698r0b-ATA8-APT.pdf
pin 32 is marked as "obsolete (was defined as IOCS16 in ATA-2, ANSI
X3.279-1996)"

> Oddly enough, that was deprecated 16 years ago, but we still use the
> feature, and we're reading a now 'unallocated' pin, which may be high or
> low. As a result, it fluctuates, but is usually set to mean 16-bit
> transfers.

If it is /IOCS15, it an open collector output from the drive which is pulled
high by R46 denoting 8 bit transfers unless the drive pulls it low. Which
would explain why the IDENTIFY command is reading the disc model number
l.i.k.e t.h.i.s with missing bytes if the drive isn't doing this.

If you're feeling brave and don't mind a bit of hacking it should be
possible to fix this. Remove R46 and install a 10K resistor between pin 32
of the IDE connector and ground. In the gap between the floppy and IDE
connectors there are four resistors. With the serial/parallel/etc sockets
pointing away from you, from left to right these are:
R14 (gap) R40 R46 R57 then C38 between IC12 and the IDE connector.

That'll force the IDE latches to always operate in 16 bit mode. The latches
being active when they shouldn't /might/ cause everything to go horribly
wrong (meaning ADFS reads the wrong data or the machine getting confused) so
I'd remove any hard drives that have data you want to keep on them if you
try this, at least until you're satisfied it works. I think there's a good
chance of it working though.

> However, if it correctly does identify the drive in 32-bit transfer
> mode, or worse still you move a drive across which has been
> pre-formatted, you can get a drive which works intermittently, and then
> causes data corruption when it flips back to 16-bit mode mid-transfer. Erk.

Thankfully the pullup (as the ATA-1 specification demands) forces it into 8
bit mode only. Which at least is consistently wrong :-)

> Newer interfaces (ie the third-party ones), don't do this - they will
> happily correctly work with the 60Gb Deskstar drive I've got. What I
> cannot tell is whether this is a hardware issue (because of the
> motherboard chipset,) or whether it's due to behaviour of ADFS itself.

The above suggests it's a hardware issue: it's how the FDC37C665GT I/O
controller was designed to be wired up. Though it's only slightly involved
with running the motherboard IDE bus - only generates a few control signals
- while most of the bus is done in discrete logic on the motherboard.

If anyone has a go I'd be interested to know if it works.

Theo

--
Theo Markettos th...@markettos.org.uk
Clare Hall, Cambridge at...@cam.ac.uk
CB3 9AL, UK http://www.markettos.org.uk/

Paul Bunyan

unread,
May 21, 2006, 6:05:23 AM5/21/06
to
"druck" <ne...@druck.freeuk.com> wrote:
> On 20 May 2006 Dave Symes <d...@ukgateway.net> wrote:
>> Unfortunately Hform 2.56 will not finish the Formatting process.
>
> Somewhere in the process you need to select LBA mode, otherwise the shape
> will be wrong. HFROM *should* still format it, but it be the wrong size
> (probably 8GB).

I had a similar rant 30/04/06 which went something like:

** begin paste **
As per label, and as !HForm (2.56) and Seagate DiscWizard (PC util) reports:

16383 cyl, 16 heads, 63 sectors: 8.46GB

As per label and as Seagate DiscWizard reports:

LBA 16,841,664: 8.62GB

!HForm reads the LBA and total capacity then tries to adust the number of
cylinders accordingly, and it all goes horribly wrong from there.
** end paste **

The conclusion at that time was !HForm was the problem rather than the
drive. Not having a Unipod or any other IDE interface means I am stuck using
small old hard disks.

Paul B.


druck

unread,
May 21, 2006, 6:03:19 AM5/21/06
to
On 21 May 2006 Dave Symes <d...@ukgateway.net> wrote:
> In article <38022f2a...@druck.freeuk.net>,
> druck <ne...@druck.freeuk.com> wrote:
> > On 20 May 2006 Dave Symes <d...@ukgateway.net> wrote:
> > > Unfortunately Hform 2.56 will not finish the Formatting process.
>
> > Somewhere in the process you need to select LBA mode, otherwise the shape
> > will be wrong. HFROM *should* still format it, but it be the wrong size
> > (probably 8GB).
>
> Thanks for the info Mr D.
> I guess what you are really saying is, she's bought a bum HD type as far
> as RO is concerned.

No it isn't, but after your last outburst when I replied to a question, I'll
leave it with that, and check why my filtering didn't work.

Brian Carroll

unread,
May 21, 2006, 6:40:48 AM5/21/06
to
In article <4e2a64...@ukgateway.net>, Dave Symes
<d...@ukgateway.net> wrote:

[Snip]

> Andrew, Thanks for that run down, I did understand what you've
> written.

[Snip]

> She does have a Unipod on the machine, with USB and Networking
> active... I suppose I'd better get the IDE update for her and
> use that as the interface.

Very interesting discussion; the sort that makes skimming the
endless dross on newsgroups worthwhile :-)

> PS: Brings forth another question, what do you use to format a
> HD connected to the Unipod IDE interface?

You can use the utility !IDETool from Simtec, which is somewhat
idiosyncratic like their !Snafu utility which you will have
probably used with the USB part of your wife's UNIpod.
Alternatively you can just keep on using !HForm with which you
will be extremely familiar but maybe with the problems already
discussed.

There is an important difference which I reported as follows to
STD last December, I'm not sure whether anything has been done
about it:

---------------------------quote---------------------------

I have just discovered today that !IDETool provides an ID length
of only 18 bits when formatting a 'bigmap', whereas the newer
versions of !HForm use 19 bits. If you study the extracts below
from the !SICK v1.27 results for my machine (which has 2
identical '80Gb' discs on a UNIpod) you will see that on
IDEFS::5 the smallest object is nearly twice the size of
IDEFS::4. This leads to a great deal of wasted disc space on
the former.

IDEFS::4 - formatted by !HForm v2.56
Disc name: Gromit
Disc filetype: &FCD (harddisc image)
Density: hard disc
Disc size: 76.34 Gbytes
RISC OS partition > 512 MB: Yes
Sector size: 512 bytes
Sectors per track: 63
Heads: 16
idlen: 19 bits (big map)
Number of zones: 2463
Size of map: 1.202636719 Mbytes
Bytes per map bit (LFAU): 8192
Smallest object: 160 Kbytes
Maximum number of objects: 524285
[...]

IDEFS::5 - formatted by !IDETool v1.3
Disc name: Mirror
Disc filetype: &FCD (harddisc image)
Density: hard disc
Disc size: 72.43 Gbytes
RISC OS partition > 512 MB: Yes
Sector size: 512 bytes
Sectors per track: 63
Heads: 16
idlen: 18 bits (big map)
Number of zones: 1169
Size of map: 584.5 Kbytes
Bytes per map bit (LFAU): 16384
Smallest object: 304 Kbytes
Maximum number of objects: 262141
[...]

The solution is to partition the drive into smaller sections or
temporarily put the drive back onto the motherboard connection
so you will have access to ADFS and !HForm.

---------------------------unquote--------------------------

Brian.

--
______________________________________________________________

Brian Carroll, Ripon, North Yorkshire, UK bric at f2s dot com
______________________________________________________________

Dave Symes

unread,
May 21, 2006, 6:54:10 AM5/21/06
to
In article <3e6c702a...@druck.freeuk.net>,

druck <ne...@druck.freeuk.com> wrote:
> On 21 May 2006 Dave Symes <d...@ukgateway.net> wrote:
> > In article <38022f2a...@druck.freeuk.net>,
> > druck <ne...@druck.freeuk.com> wrote:
> > > On 20 May 2006 Dave Symes <d...@ukgateway.net> wrote:
> > > > Unfortunately Hform 2.56 will not finish the Formatting process.
> >
> > > Somewhere in the process you need to select LBA mode, otherwise the
> > > shape will be wrong. HFROM *should* still format it, but it be the
> > > wrong size (probably 8GB).
> >
> > Thanks for the info Mr D. I guess what you are really saying is, she's
> > bought a bum HD type as far as RO is concerned.

> No it isn't, but after your last outburst when I replied to a question,
> I'll leave it with that, and check why my filtering didn't work.

> ---druck

Eh! What outburst was that!
I'm always having outbursts of one sort or another. ;-) I'm a passionate
person, it's all froth.

Dave S

--

Andrew Hill

unread,
May 21, 2006, 7:14:45 AM5/21/06
to
Theo Markettos wrote:
> Andrew Hill <us...@example.net> wrote:
>> I had a long chat with the Hitachi (well, IBM as-was) firmware team
>> about the problem, who were extremely friendly.
>>
>> To demonstrate the age of our equipment, the ADFS bus is using a legacy
>> feature from before the ATA-1 specification (that spec was written
>> 1990ish IIRC, don't quote me; pretty much BC in hard disc terms anyway,
>> and significantly pre-dating the RISC PC!) The feature determines
>> whether to perform 16-bit, or 32-bit transfers. If the RISC PC can't
>> detect the difference, it uses 16-bit transfers.
>
> Hmm... interesting. I'm not disbelieving you, but that seems a little odd
> because ATA-1 and standards until much later (mid-90s) didn't support 32 bit
> transfers. Are you sure it wasn't 8 bit transfers? There's an IOCS16 pin
> which is used to denote 16 bit transfers: some early drives (20MB, 40MB)
> could be talked to in 8 bit mode which was designed for PC XTs and similar.
> IDE was designed for 16 bit machines so no-one considered a 32 bit
> implementation until sometime later.

Yes, sorry that's what I meant :-(. It was about three years ago now
that I looked at this. I blame the dementia ;-). At least it pointed
someone with a better memory in the right direction ;-).

> ATA-1 is here:
> http://t13.org/project/d0791r4c-ATA-1.pdf
>
> /IOCS16 is an output from the drive, pin 32 on the IDE bus, which is
> connected to /IOCS16 input (pin 27) on the FDC37C665, and pulled high with
> R46, a 1K2 resistor. It's used to generate strobes to latch either 8 bit or
> 16 bit values from the drive onto the Risc PC's IO databus.

> If it is /IOCS15, it an open collector output from the drive which is pulled


> high by R46 denoting 8 bit transfers unless the drive pulls it low. Which
> would explain why the IDENTIFY command is reading the disc model number
> l.i.k.e t.h.i.s with missing bytes if the drive isn't doing this.

That's how I found it actually; initially I had presumed this was a
drive error. Then I looked at the IDENTIFY block, and found the above.
Then I asked on T13 whether anyone had any ideas, which led to a long
(taken to phone) discussion. The Hitachi guys were quite surprised that
the machine looked at this at all.

> If you're feeling brave and don't mind a bit of hacking it should be
> possible to fix this. Remove R46 and install a 10K resistor between pin 32
> of the IDE connector and ground. In the gap between the floppy and IDE
> connectors there are four resistors. With the serial/parallel/etc sockets
> pointing away from you, from left to right these are:
> R14 (gap) R40 R46 R57 then C38 between IC12 and the IDE connector.


Hmm - there's a machine upgrade service in there for anyone who wanted
it ;-). I don't think Hitachi will be the only manufacturer who will do
this in future.

>> However, if it correctly does identify the drive in 32-bit transfer
>> mode, or worse still you move a drive across which has been
>> pre-formatted, you can get a drive which works intermittently, and then
>> causes data corruption when it flips back to 16-bit mode mid-transfer. Erk.
>
> Thankfully the pullup (as the ATA-1 specification demands) forces it into 8
> bit mode only. Which at least is consistently wrong :-)

I can sadly say that wasn't 100% consistent (although I agree it is
mostly consistent). I /did/ get the drive to read in 16-bit mode once,
and then tried storing data on it. Bad stuff happens :-(, good job it
was a test drive, because I knew there were problems.

>> Newer interfaces (ie the third-party ones), don't do this - they will
>> happily correctly work with the 60Gb Deskstar drive I've got. What I
>> cannot tell is whether this is a hardware issue (because of the
>> motherboard chipset,) or whether it's due to behaviour of ADFS itself.
>
> The above suggests it's a hardware issue: it's how the FDC37C665GT I/O
> controller was designed to be wired up. Though it's only slightly involved
> with running the motherboard IDE bus - only generates a few control signals
> - while most of the bus is done in discrete logic on the motherboard.

Your additional hardware knowledge there is most helpful Theo; I've only
looked at this from a software point of view: 'Why is HForm not working
on this drive?'. 'Oooh - hang on - why is IDENTIFY DEVICE not working on
this drive?' etc etc.

Best wishes,

Drew

Andrew Hill

unread,
May 21, 2006, 8:23:35 AM5/21/06
to
Brian Carroll wrote:

--8<--- Snipped from Brian's quote from an earlier post to STD.

> The solution is to partition the drive into smaller sections or
> temporarily put the drive back onto the motherboard connection
> so you will have access to ADFS and !HForm.

--8<---

The latter is of course not recommended to Dave in this case because his
data at best would be inaccessible and would potentially be unstable ;-).

Best wishes,

Drew

Dave Plowman (News)

unread,
May 21, 2006, 11:07:04 AM5/21/06
to
In article <4e2a64...@ukgateway.net>,
Dave Symes <d...@ukgateway.net> wrote:
> PS: Brings forth another question, what do you use to format a HD
> connected to the Unipod IDE interface?

The software includes IDETool which allows formatting and partitioning. It
works well. I've got a 40 and 60 gig WD drives just bought from the local
shop and they've been fine.

--
*I brake for no apparent reason.

Brian Carroll

unread,
May 21, 2006, 11:19:40 AM5/21/06
to
In article
<44705bc6$0$2692$ed26...@ptn-nntp-reader01.plus.net>, Andrew

> --8<---

Well yes; but I did put a caveat in my posting. I understood
your explanation to imply that the problem would be
intermittent. That would mean that *if* the !HForm formatting
succeeded the drive would then be reliable for data transfer
when put back on the UNIpod IDE interface.

Tim

unread,
May 21, 2006, 11:40:19 AM5/21/06
to
In message <4e2a0e2...@davenoise.co.uk>

"Dave Plowman (News)" <da...@davenoise.co.uk> wrote:

OK so I have the top off and the first thing that strikes me is that it is
full of dust accumulated over the last 5 or more years.

How can I safely deal with that? Or do I leave well enuf alone (which seems
cowardly). Hoover nozzle maybe...?

Regards

Tim

--

Paul Bunyan

unread,
May 21, 2006, 12:59:21 PM5/21/06
to
"Tim" <T...@lynndene.demon.co.uk> wrote

> OK so I have the top off and the first thing that strikes me is that it is
> full of dust accumulated over the last 5 or more years.
>
> How can I safely deal with that? Or do I leave well enuf alone (which
> seems
> cowardly). Hoover nozzle maybe...?
>
> Regards
>
> Tim

The nozzle of a vacuum cleaner is likely to cause damage. Even the
attachments with brushes were not designed with anything as delicate as a
PCB in mind.

I'd go for a small *soft* paintbrush. About 1/2". Emphasis on *soft*. New if
poss. White spirit and dry gloss paint do nothing for bristle softness ;-)

Gently brush the the dust off whilst holding the end of the hoover nozzle
nearby to hopefully remove the dust.

Best of luck

Paul B.

PS: I have had no time this weekend to create the RisPC Hard Drive Change
page... It will be done at some point.


Andrew Hill

unread,
May 21, 2006, 1:19:27 PM5/21/06
to
Brian Carroll wrote:
> In article
> <44705bc6$0$2692$ed26...@ptn-nntp-reader01.plus.net>, Andrew
> Hill <us...@example.net> wrote:
>> Brian Carroll wrote:
>
>> --8<--- Snipped from Brian's quote from an earlier post to STD.
>
>>> The solution is to partition the drive into smaller
>>> sections or temporarily put the drive back onto the
>>> motherboard connection so you will have access to ADFS and
>>> !HForm.
>
>> --8<---
>
>> The latter is of course not recommended to Dave in this case
>> because his data at best would be inaccessible and would
>> potentially be unstable ;-).
>
> Well yes; but I did put a caveat in my posting. I understood
> your explanation to imply that the problem would be
> intermittent. That would mean that *if* the !HForm formatting
> succeeded the drive would then be reliable for data transfer
> when put back on the UNIpod IDE interface.

Aye; I was joking ;-).

Best wishes,

Drew

Ian K (N)

unread,
May 21, 2006, 2:18:45 PM5/21/06
to
In article <0a478f2...@lynndene.demon.co.uk>,

A note of warning.

If you are going to use a Hoover to clear dust from a PCB, be careful as
some can generate quite a sizable static charge, enough to kill the PCB. I
am not aware of any definitive way of preventing this other than using one
that is properly protected though.

Regards
Ian K

--
<|><|><|><|><|><|><|><|><|><|><|><|><|><|><|><|><|><|><|><|><|><|><|><|><|>

Ian karley Photocopier, Printer and Fax engineer in Dorset UK.
ne...@iank.org.uk RISC OS user and proud of it

druck

unread,
May 21, 2006, 2:02:38 PM5/21/06
to
On 21 May 2006 Tim <T...@lynndene.demon.co.uk> wrote:
> OK so I have the top off and the first thing that strikes me is that it is
> full of dust accumulated over the last 5 or more years.
>
> How can I safely deal with that? Or do I leave well enuf alone (which seems
> cowardly). Hoover nozzle maybe...?

You'll find the PCB will be a lot cleaner after using a domesic hoover, none
of those pesky little blob things clutting it up anymore.

What I've used is one of those little battery powered keyboard vacum
cleaners, which are useless for keyboards as they dont have enough suction
power to shift the discgusting gunk down there between the keys. It is just
powerful enough to remove dust as its the soft brush attachment is gently
moved over a PCB.

Dave Symes

unread,
May 21, 2006, 3:36:48 PM5/21/06
to
In article <4e2a8c3...@davenoise.co.uk>,

Dave Plowman (News) <da...@davenoise.co.uk> wrote:
> In article <4e2a64...@ukgateway.net>,
> Dave Symes <d...@ukgateway.net> wrote:
> > PS: Brings forth another question, what do you use to format a HD
> > connected to the Unipod IDE interface?

> The software includes IDETool which allows formatting and partitioning.
> It works well. I've got a 40 and 60 gig WD drives just bought from the
> local shop and they've been fine.

Thanks for that, I've had a look at te Unipod directory and it does indeed
contain IDETools.
All she needs now is the Unipod IDE upgrade.

Dave S

--

Dave Symes

unread,
May 21, 2006, 3:43:03 PM5/21/06
to
In article <4470a11e$0$18262$ed26...@ptn-nntp-reader01.plus.net>,

> Best wishes,
> Drew


Hohoho! Andrew.
If I was playing with my own machine I wouldn't particularly bother if I
broke it or not... I often do, but as this is Her machine, I want to get
it right.

So just out of interest, or fear of her wrath... Can a drive formatted by
the Unipod IDETools be read if connected to the ADFS HD interface, or the
other way around... Can an ADFS formatted disk be read if connected to the
Unipod IDE?

Cheers
Dave S

--

S G

unread,
May 21, 2006, 3:50:13 PM5/21/06
to
On 21 May Tim wrote:

> I have the top off and the first thing that strikes me is that
> it is full of dust accumulated over the last 5 or more years.
> How can I safely deal with that?

If you have a a vac. with a hose, put it on blow and go over it
with that but be sure to blow the dust out of the hose 1st (this
is a job for outside). A paint brush also helps.

--
Stewart Goldwater
http://www.janus.freeserve.co.uk/

Brian Carroll

unread,
May 21, 2006, 4:28:29 PM5/21/06
to
In article <4e2aa5...@ukgateway.net>, Dave Symes

[Snip]

> So just out of interest, or fear of her wrath... Can a drive
> formatted by the Unipod IDETools be read if connected to the
> ADFS HD interface, or the other way around... Can an ADFS
> formatted disk be read if connected to the Unipod IDE?

Yes and Yes.

It was only by chance when messing about with !SICK that I
noticed the ID length difference I wrote about in
<4e2a73dac4...@argonet.co.uk> . By then the ADFS/HForm
disc had been in use for ages on an IDEFS/IDETools UNIpod

Dave Symes

unread,
May 21, 2006, 4:48:47 PM5/21/06
to
In article <4e2aa9a89c...@argonet.co.uk>,

Brian Carroll <bric-...@argonet.co.uk> wrote:
> In article <4e2aa5...@ukgateway.net>, Dave Symes

> [Snip]

> > So just out of interest, or fear of her wrath... Can a drive
> > formatted by the Unipod IDETools be read if connected to the
> > ADFS HD interface, or the other way around... Can an ADFS
> > formatted disk be read if connected to the Unipod IDE?

> Yes and Yes.

> It was only by chance when messing about with !SICK that I
> noticed the ID length difference I wrote about in
> <4e2a73dac4...@argonet.co.uk> . By then the ADFS/HForm
> disc had been in use for ages on an IDEFS/IDETools UNIpod
> interface.

> Brian.

Thank Brian.

Dave S

--

Andrew Hill

unread,
May 21, 2006, 5:40:29 PM5/21/06
to
Dave Symes wrote:
> In article <4470a11e$0$18262$ed26...@ptn-nntp-reader01.plus.net>,
> Andrew Hill <us...@example.net> wrote:

[snippety snip poor sense of humour]

>> Aye; I was joking ;-).
>
>> Best wishes,
>> Drew
>
>
> Hohoho! Andrew.

Yes. Not quite ready for standup yet, I'm afraid :-(. Must get out more,
and stop playing with damn hard drives. I've heard that there are more
exciting things in the Big Wide World ;-). Before you ask, yes I
discovered women (getting married to one), which is probably why I'm
back to playing with computers ;-). I'm considering 'going grumpy' - any
advice?

> If I was playing with my own machine I wouldn't particularly bother if I
> broke it or not... I often do, but as this is Her machine, I want to get
> it right.

Not fun if you can't break it :-(.

> So just out of interest, or fear of her wrath... Can a drive formatted by
> the Unipod IDETools be read if connected to the ADFS HD interface, or the
> other way around... Can an ADFS formatted disk be read if connected to the
> Unipod IDE?

Yes you can. The issue was that IDETool in certain circumstances may not
format as efficiently. That probably isn't a huge issue to SWMBO, unless
she measures your idlen on a regular basis ;-).

Also works the other way round. However, IDETool supports partitions,
which ADFS doesn't, so of course don't use partitions if you want to
swap them back. Not an issue on that Hitachi drive of course, but worth
considering for the others.


One other little gotcha: RISC OS often will spend a disproportionate
amount of time looking for hard discs if there is nothing on the ADFS
bus. If the machine sits and waits for 30 seconds or so before starting
up, consider putting an old drive on the ADFS bus just so there's
something for it to find. IIRC that's a 'legacy' feature for those
drives that took ages to fire up.

Best wishes,

Drew

Dave Symes

unread,
May 22, 2006, 2:03:32 AM5/22/06
to
In article <4470de4c$0$548$ed26...@ptn-nntp-reader01.plus.net>,
Andrew Hill <us...@example.net> wrote:
> Dave Symes wrote:

[Snip]

> >
> > Hohoho! Andrew.

> Yes. Not quite ready for standup yet, I'm afraid :-(. Must get out more,
> and stop playing with damn hard drives. I've heard that there are more
> exciting things in the Big Wide World ;-). Before you ask, yes I
> discovered women (getting married to one), which is probably why I'm
> back to playing with computers ;-). I'm considering 'going grumpy' - any
> advice?

Keep up the humour practice, it gets worse with age.

Ooer! Discovered the, I'm getting hitched to her kind of woman, you are
doomed my boy... Aye we're doomed, we're all doomed Capt Mainwaring.

I've recently celebrated having the same one for 30 years, and I wouldn't
want to go back to being single again.

Andrew, unless you are like me, born grumpy, I believe it's something you
grow into, y'know, like that old pair of trousers that become so
comfortable, you never want to wear any other...
<Wife> "You think I'm going to go out shopping with you wearing those
despicable things..."

> > If I was playing with my own machine I wouldn't particularly bother if
> > I broke it or not... I often do, but as this is Her machine, I want to
> > get it right.

> Not fun if you can't break it :-(.

Indeedy, kind of spoils the exercise... Though I did manage to do a little
break on hers yesterday... Unfortunately, she was the one to find it on
the next reboot.

> > So just out of interest, or fear of her wrath... Can a drive formatted
> > by the Unipod IDETools be read if connected to the ADFS HD interface,
> > or the other way around... Can an ADFS formatted disk be read if
> > connected to the Unipod IDE?

> Yes you can. The issue was that IDETool in certain circumstances may not
> format as efficiently. That probably isn't a huge issue to SWMBO, unless
> she measures your idlen on a regular basis ;-).

> Also works the other way round. However, IDETool supports partitions,
> which ADFS doesn't, so of course don't use partitions if you want to
> swap them back. Not an issue on that Hitachi drive of course, but worth
> considering for the others.

Indeed, no partitioning will be required, she's only using it for a backup
drive anyway.

> One other little gotcha: RISC OS often will spend a disproportionate
> amount of time looking for hard discs if there is nothing on the ADFS
> bus. If the machine sits and waits for 30 seconds or so before starting
> up, consider putting an old drive on the ADFS bus just so there's
> something for it to find. IIRC that's a 'legacy' feature for those
> drives that took ages to fire up.

> Best wishes,
> Drew

The Boot drive 4 will remain on the ADFS interface.

While I had the machine completely stripped, I took the opportunity to get
the crud out of it, and also move the Boot drive from that annoying place
in the basement, up to the top floor of the second slice.

I did that years ago with this machine, it makes changing HD so much
easier... being just over 10 years old, and I always have at least two HDs
in a machine, it has seen a few HDs in that time.

Thanks for the advice, much appreciated.
I'll report back when the deed is done.

Cheers
Dave S

--

Dave Plowman (News)

unread,
May 22, 2006, 3:13:35 AM5/22/06
to
In article <4470de4c$0$548$ed26...@ptn-nntp-reader01.plus.net>,

Interesting - I have everything on a Unipod and noticed this effect, but
didn't know the cause.

Pity there wasn't a software fix as I've no room for another HD.

--
*Artificial Intelligence is no match for Natural Stupidity *

John Devine

unread,
May 22, 2006, 5:24:03 AM5/22/06
to
In article <4e2ae4b...@davenoise.co.uk>,

Dave Plowman (News) <da...@davenoise.co.uk> wrote:
> In article <4470de4c$0$548$ed26...@ptn-nntp-reader01.plus.net>,
> Andrew Hill <us...@example.net> wrote:
> > One other little gotcha: RISC OS often will spend a
> > disproportionate amount of time looking for hard discs if
> > there is nothing on the ADFS bus. If the machine sits and
> > waits for 30 seconds or so before starting up, consider
> > putting an old drive on the ADFS bus just so there's
> > something for it to find. IIRC that's a 'legacy' feature for
> > those drives that took ages to fire up.

> Interesting - I have everything on a Unipod and noticed this
> effect, but didn't know the cause.

> Pity there wasn't a software fix as I've no room for another
> HD.

It's a long time since I used my Risc PC so I could be talking
cobblers here, but can't you just configure IDEDiscs to zero?

As far as I remember that only applies to the motherboard ADFS
discs so shouldn't affect those in the UNIpod.

JD

--
John Devine Bolton Lancashire UK
e-mail: john[DOT]devine970[AT]ntlworld[DOT]com

Chris Johns

unread,
May 22, 2006, 6:06:53 AM5/22/06
to
On Sun, 21 May 2006, Andrew Hill wrote:

> One other little gotcha: RISC OS often will spend a disproportionate
> amount of time looking for hard discs if there is nothing on the ADFS
> bus. If the machine sits and waits for 30 seconds or so before starting
> up, consider putting an old drive on the ADFS bus just so there's
> something for it to find. IIRC that's a 'legacy' feature for those
> drives that took ages to fire up.

Assuming you have remembered to *configure IDEDiscs 0, that's more likely
to be CDFSSoftATAPI looking for a CD drive, it seems to take an age to
realise that there is no CD drive connected when there's nothing on the
bus.

To solve it you can put the CD drive back on the the internal bus, or
unplug the module.

If you run Select (as I do) you will may into problems with *unplug, as
the module is in a different place on the 4.02 ROMs and the softloaded
image.

Since the UNIpod can have extra modules put into it, you can get around
this with a dummy CDFSSoftATAPI module on the unipod with a higher version
number. This will stop the normal module being used, and so it won't spend
an age looking for a CD drive.

Cheers

Chris
--
Chris Johns

Brian Carroll

unread,
May 22, 2006, 5:57:24 AM5/22/06
to
In article <4e2aee6...@NoSpam-SeeSig.invalid>, John Devine
<Jo...@NoSpam-SeeSig.invalid> wrote:

[Snip]

> It's a long time since I used my Risc PC so I could be talking
> cobblers here, but can't you just configure IDEDiscs to zero?

> As far as I remember that only applies to the motherboard ADFS
> discs so shouldn't affect those in the UNIpod.

Yes, that is the answer to Andrew's comment about the delay.

One of the very confusing things about RISC OS, I''ve always
thought, is the way Config. uses 'IDEDiscs' to refer exclusively
to IDE discs on the motherboard ADFS interface (see *Status).

Likewise 'Drive 4|5' refers only to drives on that same
interface. In other words, you cannot configure IDEFS on the
UNIPod (or similar) interface using that bit of CMOS storage.

Both ADFS and IDEFS allocate Drive 4 to Master and Drive 5 to
Slave. The former boots from the drive configured in 'Drive',
but in the latter case if you want to boot from Drive 5 you have
to make Drive 4 non-bootable; ie set it to *Opt 2,0 , so it will
be 'passed over' during the boot sequence.

I don't know whether anyone really wanted to know all that :-)

Ray Dawson

unread,
May 22, 2006, 7:43:14 AM5/22/06
to
S G <nws...@ntlworld.com> wrote:

> On 21 May Tim wrote:
>
> > I have the top off and the first thing that strikes me is that it is
> > full of dust accumulated over the last 5 or more years. How can I
> > safely deal with that?
>
> If you have a a vac. with a hose, put it on blow and go over it with
> that but be sure to blow the dust out of the hose 1st (this is a job for
> outside). A paint brush also helps.
>

What I usually do is to hold my hand over the sucking end of the hoover
pipe and insert a bendy drinking straw between my fingers. This gives a
mini nozzle which is more directional and can get into the smaller crooks
and nannies.

Ray D

Timothy Hartley

unread,
May 22, 2006, 8:56:41 AM5/22/06
to
In message <gemini.izo0k10...@magray.freeserve.co.uk>
Ray Dawson <r...@magray.freeserve.co.uk> wrote:

>This gives a mini nozzle which is more directional and can get into
>the smaller crooks and nannies.

What do you do with the larger crooks and nannies or should we not
encourage any more perverted suggestions?


Tim Hartley

Dave Plowman (News)

unread,
May 22, 2006, 1:15:55 PM5/22/06
to
In article <Pine.LNX.4.58.06...@charlotte.local>,

Chris Johns <c...@eh.org.invalid> wrote:
> > One other little gotcha: RISC OS often will spend a disproportionate
> > amount of time looking for hard discs if there is nothing on the ADFS
> > bus. If the machine sits and waits for 30 seconds or so before starting
> > up, consider putting an old drive on the ADFS bus just so there's
> > something for it to find. IIRC that's a 'legacy' feature for those
> > drives that took ages to fire up.

> Assuming you have remembered to *configure IDEDiscs 0, that's more likely
> to be CDFSSoftATAPI looking for a CD drive, it seems to take an age to
> realise that there is no CD drive connected when there's nothing on the
> bus.

I've already got IDEDiscs at 0 and CDFSsoftATAPI unplugged. As I'm using
the Unipod for 2HDs and a CDR and CDW. But the machine still hangs about
on a cold boot.

> To solve it you can put the CD drive back on the the internal bus, or
> unplug the module.

> If you run Select (as I do) you will may into problems with *unplug, as
> the module is in a different place on the 4.02 ROMs and the softloaded
> image.

> Since the UNIpod can have extra modules put into it, you can get around
> this with a dummy CDFSSoftATAPI module on the unipod with a higher
> version number. This will stop the normal module being used, and so it
> won't spend an age looking for a CD drive.

--
*No husband has ever been shot while doing the dishes *

John Kortink

unread,
May 22, 2006, 2:41:04 PM5/22/06
to

And in a thread entitled 'New Hard Drive' no less.

Stop it now, before people get hurt.


John Kortink

--

Email : kor...@inter.nl.net
Homepage : http://www.inter.nl.net/users/J.Kortink

Tim

unread,
May 22, 2006, 4:46:20 PM5/22/06
to
In message <0a478f2...@lynndene.demon.co.uk>
Tim <T...@lynndene.demon.co.uk> wrote:

[snip]

Frabjous day!

I now have a new hard drive! And its so quiet - none of that clicking and
whittering....

Now the next task - replacing the ROMS to move from 4.02 to 4.39!

However, I am leaving that for a while to enjoy the quiet satisfaction and
let some of the hair that went white return to normal.

Many many thanks to all who offered advice.

Regards

--
Tim

Andrew Hill

unread,
May 22, 2006, 6:26:53 PM5/22/06
to
Dave Plowman (News) wrote:
> In article <Pine.LNX.4.58.06...@charlotte.local>,
> Chris Johns <c...@eh.org.invalid> wrote:
>>> One other little gotcha: RISC OS often will spend a disproportionate
>>> amount of time looking for hard discs if there is nothing on the ADFS
>>> bus. If the machine sits and waits for 30 seconds or so before starting
>>> up, consider putting an old drive on the ADFS bus just so there's
>>> something for it to find. IIRC that's a 'legacy' feature for those
>>> drives that took ages to fire up.
>
>> Assuming you have remembered to *configure IDEDiscs 0, that's more likely
>> to be CDFSSoftATAPI looking for a CD drive, it seems to take an age to
>> realise that there is no CD drive connected when there's nothing on the
>> bus.
>
> I've already got IDEDiscs at 0 and CDFSsoftATAPI unplugged. As I'm using
> the Unipod for 2HDs and a CDR and CDW. But the machine still hangs about
> on a cold boot.

Indeed. That's how my machine is too; I have tried the obvious stuff
(IDEDiscs 0, Unplug CDFSSoftPower) to no success.

Eventually I got pee-ed off and *Unplugged ADFS and ADFSFiler. Drastic,
but does work ;-). The downsides to that should be fairly obvious (no
floppy drive).

Best wishes,

Drew

John Cartmell

unread,
May 22, 2006, 7:39:26 PM5/22/06
to
In article <34212f2...@lynndene.demon.co.uk>,

Tim <T...@lynndene.demon.co.uk> wrote:
> However, I am leaving that for a while to enjoy the quiet satisfaction and
> let some of the hair that went white return to normal.

So when do you want the bad news ...

? ;-)

--
John Cartmell john@ followed by finnybank.com 0845 006 8822
Qercus magazine FAX +44 (0)8700-519-527 www.finnybank.com
Qercus - the best guide to RISC OS computing

Dave Plowman (News)

unread,
May 23, 2006, 4:47:03 AM5/23/06
to
In article <44723aac$0$18226$ed26...@ptn-nntp-reader01.plus.net>,

Andrew Hill <us...@example.net> wrote:
> Indeed. That's how my machine is too; I have tried the obvious stuff
> (IDEDiscs 0, Unplug CDFSSoftPower) to no success.

> Eventually I got pee-ed off and *Unplugged ADFS and ADFSFiler. Drastic,
> but does work ;-). The downsides to that should be fairly obvious (no
> floppy drive).

Ah. I'll give that a try. My floppy is rarely used.

--
*Nothing is foolproof to a sufficiently talented fool*

VinceH

unread,
May 23, 2006, 12:21:34 PM5/23/06
to
On 23 May 2006, Dave Plowman (News) wrote:

> My floppy is rarely used.

I'm not at all sure we needed to know that! =:-o

;-)

--
VinceH - http://www.softrock.co.uk/info/vinceh.html

Andrew Hill

unread,
May 23, 2006, 1:47:39 PM5/23/06
to
VinceH wrote:
> On 23 May 2006, Dave Plowman (News) wrote:
>
>> My floppy is rarely used.
>
> I'm not at all sure we needed to know that! =:-o
>
> ;-)
>
Oh no, it's OK. It's only because he's replaced it with a big hard drive
:-O.

The old 'uns are still the good 'uns :-D.

Best wishes,

Drew

John Kortink

unread,
May 23, 2006, 2:51:06 PM5/23/06
to
On Tue, 23 May 2006 16:21:34 GMT, VinceH <sp...@softrock.co.uk> wrote:

>On 23 May 2006, Dave Plowman (News) wrote:
>
>> My floppy is rarely used.
>
>I'm not at all sure we needed to know that! =:-o
>
>;-)

I'm guessing a woman must be involved in this case,
'cause most don't appreciate floppies.

Tim

unread,
May 23, 2006, 3:35:46 PM5/23/06
to
In message <4e2b3ef...@cartmell.demon.co.uk>
John Cartmell <jo...@cartmell.demon.co.uk> wrote:

> In article <34212f2...@lynndene.demon.co.uk>,
> Tim <T...@lynndene.demon.co.uk> wrote:
> > However, I am leaving that for a while to enjoy the quiet satisfaction and
> > let some of the hair that went white return to normal.
>
> So when do you want the bad news ...
>
> ? ;-)
>

About the hair or the ROMs?

As for the latter I notice that I will have to take out the network card -
sigh...

--

Mark Rowan

unread,
May 23, 2006, 3:56:45 PM5/23/06
to
In article <v5m6721b5oakmgsn2...@4ax.com>,

John Kortink <kor...@inter.nl.net> wrote:
> On Tue, 23 May 2006 16:21:34 GMT, VinceH <sp...@softrock.co.uk> wrote:
> > On 23 May 2006, Dave Plowman (News) wrote:
> > > My floppy is rarely used.

> > I'm not at all sure we needed to know that! =:-o

> > ;-)

> I'm guessing a woman must be involved in this case,
> 'cause most don't appreciate floppies.

Bit of advice: if the floppy is ejected, you'll have problems closing any
ZIP you may currently have open, and may come up against truncation
errors. If this happens, you should reinsert the floppy and try again.

--
If you don't care where you are, you're not lost.

||Mark Rowan|| Iyonix PC, 128MB, 80GB, RO 5.11 | Skype tamias_striatus
http://www.tamias.co.uk | mark(at)tamias co uk | ICQ 30759398 | ZFC-A(Ca)
Reply-to is valid, PGP mail preferred: http://tamias.co.uk/cgi-bin/email

John Kortink

unread,
May 23, 2006, 6:05:38 PM5/23/06
to
On Tue, 23 May 2006 20:56:45 +0100, Mark Rowan <arti...@hotpop.com>
wrote:

>In article <v5m6721b5oakmgsn2...@4ax.com>,
> John Kortink <kor...@inter.nl.net> wrote:
>> On Tue, 23 May 2006 16:21:34 GMT, VinceH <sp...@softrock.co.uk> wrote:
>> > On 23 May 2006, Dave Plowman (News) wrote:
>> > > My floppy is rarely used.
>
>> > I'm not at all sure we needed to know that! =:-o
>
>> > ;-)
>
>> I'm guessing a woman must be involved in this case,
>> 'cause most don't appreciate floppies.
>
>Bit of advice: if the floppy is ejected, you'll have problems closing any
>ZIP you may currently have open, and may come up against truncation
>errors. If this happens, you should reinsert the floppy and try again.

I see.

I never looked at it from that perspective.

Timothy Hartley

unread,
May 24, 2006, 4:30:18 AM5/24/06
to
In message <mn1772pub1u6k1tgb...@4ax.com>
John Kortink <kor...@inter.nl.net> wrote:

> On Tue, 23 May 2006 20:56:45 +0100, Mark Rowan <arti...@hotpop.com>
> wrote:
>
>>In article <v5m6721b5oakmgsn2...@4ax.com>,
>> John Kortink <kor...@inter.nl.net> wrote:
>>> On Tue, 23 May 2006 16:21:34 GMT, VinceH <sp...@softrock.co.uk> wrote:
>>> > On 23 May 2006, Dave Plowman (News) wrote:
>>> > > My floppy is rarely used.
>>
>>> > I'm not at all sure we needed to know that! =:-o
>>
>>> > ;-)
>>
>>> I'm guessing a woman must be involved in this case,
>>> 'cause most don't appreciate floppies.
>>
>>Bit of advice: if the floppy is ejected, you'll have problems closing any
>>ZIP you may currently have open, and may come up against truncation
>>errors. If this happens, you should reinsert the floppy and try again.
>
> I see.
>
> I never looked at it from that perspective.
>
>
> John Kortink

I can understand that - only a contortionst could.

Tim Hartley

Theo Markettos

unread,
May 24, 2006, 9:34:36 AM5/24/06
to
Andrew Hill <us...@example.net> wrote:
> That's how I found it actually; initially I had presumed this was a
> drive error. Then I looked at the IDENTIFY block, and found the above.
> Then I asked on T13 whether anyone had any ideas, which led to a long
> (taken to phone) discussion. The Hitachi guys were quite surprised that
> the machine looked at this at all.

Ah, right. I wondered how you managed to get in touch with the firmware
team. The T13 mailing list is quite an interesting read for
understanding/writing the internals of ATA (IDE)/ATAPI drivers. (Which I
was, once upon a time)

> Hmm - there's a machine upgrade service in there for anyone who wanted
> it ;-). I don't think Hitachi will be the only manufacturer who will do
> this in future.

:-)

> I can sadly say that wasn't 100% consistent (although I agree it is
> mostly consistent). I /did/ get the drive to read in 16-bit mode once,
> and then tried storing data on it. Bad stuff happens :-(, good job it
> was a test drive, because I knew there were problems.

That's odd. I was assuming it was unconnected on the drive: that would mean
the RPC interface was permanently in 8 bit mode. I can't see any way that
might change, unless the drive is somehow driving the line. Which is odd if
they really are treating it as deprecated. In which case a track cut and
wiring the pin to ground might be safer.

TBF the Risc PC is behaving as it was designed: in 1994 when it was
released, ATA-1 was the current standard and /IOCS16 was mandatory.

> Your additional hardware knowledge there is most helpful Theo; I've only
> looked at this from a software point of view: 'Why is HForm not working
> on this drive?'. 'Oooh - hang on - why is IDENTIFY DEVICE not working on
> this drive?' etc etc.

[grins] I probably wouldn't have thought of it if you hadn't noticed that
:-)

Theo

Dave Symes

unread,
May 31, 2006, 1:56:48 AM5/31/06
to
Just to round up...

Got the Unipod IDE function upgrade and fitted it.
Ran !Snafu & reflashed ROM
Connected her new HD to it.
Did a format job, without any problems.
She is now a happy woman.

Thank to all for the advice, particularly Andrew Hill and Stuart T.

Cheers
Dave S

--

Reply all
Reply to author
Forward
0 new messages