What do you think, will connecting a ST-225 work (part 2)
Rich
ST-225 HD on my relatively modern (1998) motherboard (MS-6119), with an 8
bit controller card (WD1002A0WX1).
I'm working totally with DOS. All IDE set to "not intalled" in my system
BIOS. Floppy disk connected.
The procedure to follow is:
1 Low level format the drive using the programme that comes with the
controller card. This is effected by runing DEBUG, and at the prompt:
G=C800:5.
2 Partition the drive using FDISK
3 Hi level format the drive using FORMAT
Report:
I can low level format the drive, but I cannot create a partition with
FDISK.
This is the point I think where some (not all) are prevented from installing
an old ST-225.
Running FDISK produces "Error Reading Fixed Disk". Running FDISK/STATUS
produces:
DISK DRV MBYTES FREE USAGE
1 20 20 %
When my PC starts up, the drive is interrogated, because I see the red LED
on the drive itself flash a few times.
If I enable IDE0 to boot from my regular hard drive, when the controller
card is in, the system promts me to put in a floppy boot disk. When the card
is out, the system boots from my regular C:
I believe the ST-225 is using BIOS in the controller card.
I think that in some cases the problem can be caused by system BIOS not
getting along with the controller BIOS.
I don't know what further diagnosis or procedure I could do to psyche
out the problem. There may be no fix I realise, not even with a 16 bit
controller card.
Motherboard: http://artofhacking.com/th99/m/M-O/34347.htm
Controller card: http://artofhacking.com/th99/c/U-Z/20217.htm
P.S. I don't know if using FIXMBR would help. Probably not.
DerekW
"Rich" <no...@emailo.com> wrote in message
news:7lkt84F...@mid.individual.net...
I reckon its going to be a struggle the older MFM HDD's had to be manually
configured
into the controller bios as the technology was not capable of reading the
drive bios (I remember spending ages searching for the spec for a very
uncommon and huge for the time 105mb Teac drive) have you got the bios
jumpers configured correctly it was a common issue prior plug and play to
get interrupt and address conflicts - still not half as awkward as the
integrated IO cards with parallel and and serial port jumpers
Derek
http://stason.org/TULARC/pc/hard-drives-hdd/seagate/ST225-21MB-5-25-HH-MFM-ST412.html
http://www.artofhacking.com/th99/c/U-Z/21202.htm
John Jordan
>
> I think that in some cases the problem can be caused by system BIOS not
> getting along with the controller BIOS.
You have marked IRQ 5 (or 2/9 if that's what the jumper is set to) and
DMA 3 as reserved for use by ISA cards, right? Apparently DMA was the
normal transfer method for XT HD controllers so even fdisk may use it.
> P.S. I don't know if using FIXMBR would help. Probably not.
Hopefully after the low-level format the whole disk is full of zeros,
which should be fine by fdisk.
--
John Jordan
DerekW
"DerekW" <nos...@ntlworld.com> wrote in message
news:pUkJm.15719$9M4....@newsfe03.ams2...
there are issues with later than 6.2 with MFM drives not
certain but I might have something a bit earlier stashed
away.
Derek
Rich
"John Jordan" <ju...@jaj22.org.uk> wrote in message
news:NzmJm.132461$R52....@newsfe24.ams2...
Hi, this could be a right nightmare and I may fail, but actually, I'm
learning about PCs doing this, and enjoying attempting this crazy thing. :c)
I was completely gobsmacked when FDISK could not read or write to the drive.
Because of course, I'd got the PC to low-level format. It *looked like* then
that: a) the ST-225 is working, b) the ISA controller card is working C)
system is all working fine.
I know that all that FDISK really does is lay down Standard MBR on the boot
sector of a disk.
My PC obviously knows there is "a disk" connected to the system. And it does
so by relying on BIOS on the controller card, not my regular AMI BIOS. What
is interesting is that FDISK /STATUS does return some data about the disk.
DISK DRV MBYTES FREE USAGE
1 20 20 %
http://artofhacking.com/th99/c/U-Z/20217.htm
Okay, card settings at presnt:
W3: closed (BIOS enabled)
W4: 2-3 (i/o 320h)
W5: 1-2 (hardwired) ROM size 32K 64K
W6: open (correct because my HD is just 4 head)
W7: 1-2 (hardwired) (Interrupt IRQ5)
W8: 2-3 (Single Controller setting, primary contoller)
s1/jumper7: open (IRQ5)
s1/jumper8: (AT mode enabled)
So, I'm IRQ5 and I've not tried to change s1/jumper7.
As to DMA: The DMA channels are 0, 3, 5, 6, 7 and all set to Pnp. The option
is Pnp or ISA / EISA
As to IRQ I see IRQ 3, 4, 5, 7, 9 and these are all set to PCI/Pnp. Options
are PCI/Pnp or ISA / EISA.
The HD controller card is in IDE slot 1.
I've never had anything in an ISA slot bfore, neither an old 8 bit card.
As to DOS verion: I've taken DOS from Windows98 First Eition, so it's
version 7.1. Any DOS programs I've coped onto a floppy I'm taking from
Windows98.
Rich
> The HD controller card is in IDE slot 1.
Errata: ISA slot 1.
Mike Tomlinson
writes
>1 Low level format the drive using the programme that comes with the
>controller card. This is effected by runing DEBUG, and at the prompt:
>G=C800:5.
Well done. What was it in the end, you or the card?
>Running FDISK produces "Error Reading Fixed Disk". Running FDISK/STATUS
>produces:
>
>DISK DRV MBYTES FREE USAGE
>
> 1 20 20 %
Hmm. No errors from a c800:5 format? Did you hear the drive tick
regularly as the format progressed? Were you asked for an interleave
value?
>When my PC starts up, the drive is interrogated, because I see the red LED
>on the drive itself flash a few times.
good.
>If I enable IDE0 to boot from my regular hard drive, when the controller
>card is in, the system promts me to put in a floppy boot disk. When the card
>is out, the system boots from my regular C:
good.
>I believe the ST-225 is using BIOS in the controller card.
must be, if you see the drive light flash as the card initialises.
>P.S. I don't know if using FIXMBR would help. Probably not.
I doubt it very much. Have you tried the other drive you say you had?
--
(\__/)
(='.'=) Bunny says Windows 7 is Vi$ta reloaded.
(")_(") http://imgs.xkcd.com/comics/windows_7.png
Mike Tomlinson
<ju...@jaj22.org.uk> writes
>Hopefully after the low-level format the whole disk is full of zeros,
Do you know whether that's defined? I've seen zeros, 0xFF, 0xF6 and
probably a myriad others that I've forgotten.
Wonder whether a Linux LiveCD would recognise the XT controller card the
OP has? :)
Rich
"Mike Tomlinson" <mi...@jasper.org.uk> wrote in message
news:QPAHX7Ie...@jasper.org.uk...
> In article <NzmJm.132461$R52....@newsfe24.ams2>, John Jordan
> <ju...@jaj22.org.uk> writes
>
>>Hopefully after the low-level format the whole disk is full of zeros,
>
> Do you know whether that's defined? I've seen zeros, 0xFF, 0xF6 and
> probably a myriad others that I've forgotten.
>
> Wonder whether a Linux LiveCD would recognise the XT controller card the
> OP has? :)
Interestingly, I have a version on CD of Vector Linux v5.8 solo. Don't know
if installing that would have mileage.
Rich
> In article <7lkt84F...@mid.individual.net>, Rich <no...@emailo.com>
> writes
>
>>1 Low level format the drive using the programme that comes with the
>>controller card. This is effected by runing DEBUG, and at the prompt:
>>G=C800:5.
>
> Well done. What was it in the end, you or the card?
>
>>Running FDISK produces "Error Reading Fixed Disk". Running FDISK/STATUS
>>produces:
>>
>>DISK DRV MBYTES FREE USAGE
>>
>> 1 20 20 %
>
> Hmm. No errors from a c800:5 format? Did you hear the drive tick
> regularly as the format progressed? Were you asked for an interleave
> value?
Yep, It sort of makes a ratchety sound (followed by a small beep)and
repeats. Goes on for hours. Interleave value 3. I did the simplified
low-level format option. I even did a dynamic low-level format with
settings:
ccc h rrr ppp ee o
615 4 621 621 11 5
I should really rely on the simplified I think.
>>When my PC starts up, the drive is interrogated, because I see the red LED
>>on the drive itself flash a few times.
>
> good.
>
>>If I enable IDE0 to boot from my regular hard drive, when the controller
>>card is in, the system promts me to put in a floppy boot disk. When the
>>card
>>is out, the system boots from my regular C:
>
> good.
>
>>I believe the ST-225 is using BIOS in the controller card.
>
> must be, if you see the drive light flash as the card initialises.
>
>>P.S. I don't know if using FIXMBR would help. Probably not.
>
> I doubt it very much. Have you tried the other drive you say you had?
Yes, a bit. I was trying not to low level-format this, because I was wanting
to see any data on it.
DOS utility MBRTool ought to read the first sector of the disk, but it only
shows ??????????... Meaning, it never accessed the disk - I bet.
I wonder if it's BIOS DMA thing. Or IRQ thing.
Mike Tomlinson
writes
>So, I'm IRQ5 and I've not tried to change s1/jumper7.
>
>As to DMA: The DMA channels are 0, 3, 5, 6, 7 and all set to Pnp. The option
>is Pnp or ISA / EISA
Would have thought it is obvious. Set DMA 3 to ISA.
>As to IRQ I see IRQ 3, 4, 5, 7, 9 and these are all set to PCI/Pnp. Options
>are PCI/Pnp or ISA / EISA.
and set IRQ5 to ISA.
>The HD controller card is in IDE slot 1.
ISA slot 1...
>I've never had anything in an ISA slot bfore, neither an old 8 bit card.
>
>As to DOS verion: I've taken DOS from Windows98 First Eition, so it's
>version 7.1. Any DOS programs I've coped onto a floppy I'm taking from
>Windows98.
Eugh. Get a copy of DOS 6.22 or DR-DOS.
Jaimie Vandenbergh
>As to DMA: The DMA channels are 0, 3, 5, 6, 7 and all set to Pnp. The option
>is Pnp or ISA / EISA
I'm not sure the card supports DMA, so this should not matter. Reserve
DMA3 for ISA/EISA anyway, as JJ recommends - it certainly won't hurt.
>As to IRQ I see IRQ 3, 4, 5, 7, 9 and these are all set to PCI/Pnp. Options
>are PCI/Pnp or ISA / EISA.
You should definitely set IRQ5 to ISA/EISA. Again this probably won't
make a difference, since 5 is generally unused by the mobo anyway.
If you haven't already, I'd sugges turning off the serial and parallel
ports in the BIOS, in case they decide to reserve strange DMA/IRQs.
The ISA slot you have it in will not matter - all ISA slots are
created equal, which is why you had to hard-code the bloody IRQ and
DMA all the time.
Cheers - Jaimie
--
"The problem is not that the world is full of fools, it's that lightning
isn't being distributed correctly." - Mark Twain
Mike Tomlinson
writes
>Yep, It sort of makes a ratchety sound (followed by a small beep)
Good, that's it finding track 0, then track 612 and going back during
its initialisation routine. It should do it just once, when the drive
is powered up.
>and
>repeats. Goes on for hours.
Not good. May mean it can't find track 0, or you've set the interleave
too low for the drive to cope.
> Interleave value 3.
Too low. Try 6 for an XT-class controller.
>ccc h rrr ppp ee o
>615 4 621 621 11 5
What's o? Maybe interleave...
>Yes, a bit. I was trying not to low level-format this, because I was wanting
>to see any data on it.
OK, fair enough.
And you still haven't told us what you did to get the shebang working in
the first place... spread your knowledge!
DerekW
If IIRC and it is a Dos version problem I have got an answer mail me at
pandairderek
at that search engine beginning with Go you will need a mail box that can
receive large files say about floppy size!
Derek
treblesix
As far as I recall from having various PCs from that era with
RLL and MFM drives, all the blasted machines had different
proprietary boot disks with various varieties of DOS and system
specific stuff. Dos varied between V3 and V5 and DRDOS
was as common as MSDOS. Just about every additional
card for memory expansion and extra drives came with a
driver/install floppy and instructions. Floppies were either
5 1/4" or 720K low density floppies. Nightmare era lol.
Perhaps this might be helpful ?????
http://www.vintage-computer.com/vcforum/archive/index.php/t-4352.html
Rich
----
After 5 hours I get:
"Format Successful
System will now restart
Insert DOS diskette in drive A:
Press any key when ready."
The ST-225, I believe, is regarded as a "standard" 20 MB HD (615 cylinders,
4 heads). On simplified formatting (which I think may be the best choice)
the interleave is 3 and it's recommended. These instructions were made mid
1980's or therabouts. I've done the simplified formating:
-----
http://www.uncreativelabs.net/textfiles/drives/20MEG.FMT
Re: dynamic formatting:
"The Super BIOS program displays the initiate dynamic configuration
message:Key in disk characteristics as follows:
ccc h rrr ppp ee o where:
ccc = total number of cylinders (1-4 digits)
h = number of heads (1-2 digits)
rrr = starting reduced write current cylinder (1-4 digits)
ppp = write precomp cylinder (1-4 digits)
ee = max correctable error burst length (1-2 digits) range
= 5 to 11 bits, default = 11 bits
o = CBB option byte, step rate select (1 hex digit) range
= 0 to 7, default = 5 refer to controller and drive
specification for step rates.
Enter each value separated by a space and follow the complete entry
with a RETURN. Enter physical cylinder values.
Example for ST 251: 820 6 821 410 11 5 (CR)
note st-225 620 4 621 621 11 s (cr)"
I think there is errata with the note above, for ST-225 probably should be:
615 4 621 621 11 5
-----
My original problem was simply a video card writing code all over C800:0000
area, and prevented DEBUG -G=C800:5 from working. Now I can low-level
format, but cannot get any read or write so cannot get FDISK to write the
standard MBR.
Rich
"Jaimie Vandenbergh" <jai...@sometimes.sessile.org> wrote in message
news:4lgdf59p2htfsci8v...@4ax.com...
I sometimes see folks talking about setting IDE0 to type 2. Mine is a type 2
drive. The controller card is ST-412 type. There is I think some relation in
type ST-412 controller cards to IDE, but I'm not sure whether having IDE0
set to type 2 can have any relavence work. I'm thinking not because, well,
I'm not suppossed, I guess, to have an IDE card or controller card
installed. But there is relationship between IDE and ST-412.
I'll make adjustments suggested.
Mike Tomlinson
writes
]
>I sometimes see folks talking about setting IDE0 to type 2. Mine is a type 2
>drive. The controller card is ST-412 type. There is I think some relation in
>type ST-412 controller cards to IDE
There is, but only in the software interface*. ST-412, a hardware
interface definition, evolved from ST-506 (they're essentially the same
thing.) If you want a similar example, look at SASI and SCSI.
http://en.wikipedia.org/wiki/ST-412
* The software interface to the controller is referred to as
"1700-compatible" or similar (I've googled for the precise spec but
can't find it, my google-fu's weak tonight.) This is in turn used by
BIOS interrupt 13h, which BIOS uses to boot.
IDE disks have to adhere to this spec if it is desired to boot from
them. Once specific OS-level drivers have been loaded it's up to those
drivers what to do with data presented to them. This is how modern
versions of Windows boot: they load using the basic BIOS interrupt, then
load their own drivers to load the rest of the OS using (often faster)
drivers more suited to the specific disk controller in use.
Mike Tomlinson
writes
>After 5 hours I get:
>
>"Format Successful
>System will now restart
>Insert DOS diskette in drive A:
>Press any key when ready."
Too long. It should take ~10 - 15 mins or so, no more.
>The ST-225, I believe, is regarded as a "standard" 20 MB HD (615 cylinders,
>4 heads).
Yes. Different sources will give you different results. My HD bible
dating back to March 1995 says
cap 21M
cyl 615
hd 4
spt 17
wpc 300
> On simplified formatting (which I think may be the best choice)
>the interleave is 3 and it's recommended.
nah.
> These instructions were made mid
>1980's or therabouts.
pah. by newbies.
> I've done the simplified formating:
>http://www.uncreativelabs.net/textfiles/drives/20MEG.FMT
Hell, I remember all that.
They are very very wrong suggesting interleave 3 for an XT controller.
If it runs at all, it'll run like a stunned slug.
>ee = max correctable error burst length (1-2 digits) range
>= 5 to 11 bits, default = 11 bits
agreed, stick with that.
>o = CBB option byte, step rate select (1 hex digit) range
>= 0 to 7, default = 5 refer to controller and drive
>specification for step rates.
Stick with 5. Any other value may cause the controller to attempt to
step the drive heads too fast, leading to problems.
>note st-225 620 4 621 621 11 s (cr)"
620 cyls is too many. That'll cause the sort of problems you have been
seeing. I recommend the following:
612 4 612 612 11 5
>I think there is errata with the note above, for ST-225 probably should be:
>
>615 4 621 621 11 5
hmm. 621 is beyond the max cylinder the drive can access. For ex, the
parking cylinder is spec'd as 621, that'll just cause the drive to
repeatedly bang the head arms into the central spindle, possibly with
damage.
>My original problem was simply a video card writing code all over C800:0000
>area, and prevented DEBUG -G=C800:5 from working. Now I can low-level
>format, but cannot get any read or write so cannot get FDISK to write the
>standard MBR.
You're almost there, don't worry.
Rich
> In article <7lojfrF...@mid.individual.net>, Rich <no...@emailo.com>
> writes
>
>>After 5 hours I get:
>>
>>"Format Successful
>>System will now restart
>>Insert DOS diskette in drive A:
>>Press any key when ready."
>
> Too long. It should take ~10 - 15 mins or so, no more.
Wow! Thing is, after the process, I do get the proper message for succesful
low-level format.
>>The ST-225, I believe, is regarded as a "standard" 20 MB HD (615
>>cylinders,
>>4 heads).
>
> Yes. Different sources will give you different results. My HD bible
> dating back to March 1995 says
>
> cap 21M
> cyl 615
> hd 4
> spt 17
> wpc 300
>
>> On simplified formatting (which I think may be the best choice)
>>the interleave is 3 and it's recommended.
>
> nah.
>
>> These instructions were made mid
>>1980's or therabouts.
>
> pah. by newbies.
>
>> I've done the simplified formating:
>>http://www.uncreativelabs.net/textfiles/drives/20MEG.FMT
>
> Hell, I remember all that.
>
> They are very very wrong suggesting interleave 3 for an XT controller.
> If it runs at all, it'll run like a stunned slug.
I'm running the old controller on "AT mode enabled". Because my system is
AT. But the WD1002A-WX1 is essentially an XT controller I guess.
>
>>ee = max correctable error burst length (1-2 digits) range
>>= 5 to 11 bits, default = 11 bits
>
> agreed, stick with that.
>
>>o = CBB option byte, step rate select (1 hex digit) range
>>= 0 to 7, default = 5 refer to controller and drive
>>specification for step rates.
>
> Stick with 5. Any other value may cause the controller to attempt to
> step the drive heads too fast, leading to problems.
>
>>note st-225 620 4 621 621 11 s (cr)"
>
> 620 cyls is too many. That'll cause the sort of problems you have been
> seeing. I recommend the following:
>
> 612 4 612 612 11 5
I thought the cylinders in the ST-225 handbook were the nature of a
parameter of the ST-225, of 512 cylinders, and that is what you always put
in whenever specifying a cylinders value in HD install programs. Seems not.
:c)
>>I think there is errata with the note above, for ST-225 probably should
>>be:
>>
>>615 4 621 621 11 5
This is why I changed the cylinders from 620 to 615.
> hmm. 621 is beyond the max cylinder the drive can access. For ex, the
> parking cylinder is spec'd as 621, that'll just cause the drive to
> repeatedly bang the head arms into the central spindle, possibly with
> damage.
>
>>My original problem was simply a video card writing code all over
>>C800:0000
>>area, and prevented DEBUG -G=C800:5 from working. Now I can low-level
>>format, but cannot get any read or write so cannot get FDISK to write the
>>standard MBR.
>
> You're almost there, don't worry.
Weird why the format takes sop long. But then again I have a very old HD,
with a very old (but compatible) HD controller, in a mobo made about 14
years after these items.
I wonder if I need IBM PC DOS 3.0. Or some other verion of DOS.
DerekW
Starwars
email me if you want to try one at pandairderek at that search engine that
starts with Go dot com-po need to take email attachment about floppy size
:-)
Derek
DerekW
John Jordan
>
> I'm running the old controller on "AT mode enabled". Because my system
> is AT. But the WD1002A-WX1 is essentially an XT controller I guess.
I'm still wondering what on earth this jumper does.
--
John Jordan
Rich
Hi. I tried yesterday to contact, but email bounced.
My email is:
Please remove capital letters (GARBAGEBIT). There is un underscore after y
in hobby.
Rich
Rich
> Thats what I was thinking in a post far far away ......oh hang on thats
> Starwars
> email me if you want to try one at pandairderek at that search engine that
> starts with Go dot com-po need to take email attachment about floppy size
> :-)
> Derek
Hi. I tried yesterday to contact, but email bounced.
My email is:
Please remove capital letters (GARBAGEBIT). There is an underscore after y
in "hobby".
Rich
Rich
> Thats what I was thinking in a post far far away ......oh hang on thats
> Starwars
> email me if you want to try one at pandairderek at that search engine that
> starts with Go dot com-po need to take email attachment about floppy size
> :-)
> Derek
Hi. I tried yesterday to contact, but email bounced.
Rich
I don't even know why a card for an XT is not (usually) compatible with an
AT machine. :c)
Some questions (for everbody):
Is it about right that it takes 5 hours to physically format the 20 meg
drive? I mean, it is setting the disk geometry.
Can you choose any value for cylinders, or do you have to go with 615 (the
handbook value)?
Anyway, because the red LED lights up on the drive on (DOS) startup, most, I
think, would have bet the drive is readable and writeable. But FDISK just
won't show it's menu, because there is no drive accessibility, all FDISK
will do is show there is a 20 meg disk installed, when the STATUS swith is
used. Clearly my system BIOS hands over (in some way) to the controller card
BIOS in order that the HD can be accessed. But I know that this arrangement
of outboard controller card, is not the modern arrangement. But I do also
know that folks have used this arrangement right up to Pentium 3. But,
coming up short at FDISK is, I think the thing that some others (like me)
get stuck on.
It's no big thing, I'm just messing about. But I wish I could crack this, by
and by. :c) But, nothing has worked yet.
Jaimie Vandenbergh
>Is it about right that it takes 5 hours to physically format the 20 meg
>drive? I mean, it is setting the disk geometry.
No. From rusty memory, it should be between 10 minutes and half an
hour.
Excessive time would normally mean (a) that your interleave is too
small - so rather than being able to jump from one cylinder to the
next and catch the start of the new cylinder, it's missing and having
to take an entire extra revolution before it gets going.
Or (b) that the disk is damaged and going through many retries to seek
or verify, but you'd be able to tell by the different clicky noises
the dodgy bits would make.
Cheers - Jaimie
--
"We all recall that the difference between a computer salesman and a car
salesman is that the car salesman *knows* he's lying to you"
"... and probably knows how to drive"
- F O'Donnell and M Smith, in afs
Rich
Under a normal situation there would be an IDE HD fitted, and it would be
accessed through one of the IDE controllers on my motherboard. My systems
CMOS (BIOS chip) needs to know the disk parameters. You tell CMOS these
parameters when you have your IDE enabled and you either have AUTO set, or
you specify a drive type, or you enter the HD details yourself. In this
situation you are enabling the PC to read and write to the drive through, of
course, the controller which happens to be integrated with the motherboard.
My situation at the moment is that IDE is disabled and the HD contoller sits
in an IDE slot.
Now, in this situation you disable IDE controllers. But, despite that, I do
read in places, about telling BIOS about the drive parameters. That is
setting CMOS with the parameters of the drive. That confuses me a bit,
because I'm thinking, if you disable IDE in BIOS, then how can you set any
drive parameters. Yet, this is exactly what is being talked about here:
http://www.vintage-computer.com/vcforum/archive/index.php/t-4352.html
Now, I have a type 2 drive, and I can choose that in my system BIOS.
There are two kinds of disabling options: You can set IDEO, IDE1, IDE2, IDE3
to "not installed" or you can simply disable IDE elsewhere in the BIOS menu.
You must be able to disable IDE, yet still set IDE0 to type 2?
I am confused here because how can you set IDE0 to type 2, if you disable
IDE0.?
So, I don't really understand that article on vintage computer forum. Can
anyone please tell me what's the scoop? Thanks.
Rich
course, the controller which happens to be integrated with the motherboard.
>
> My situation at the moment is that IDE is disabled and the HD contoller
> sits in an IDE slot.
Aaargh, I keep doing thsat, I mean ISA slot.
Adrian C
> My situation at the moment is that IDE is disabled and the HD contoller
> sits in an IDE slot.
>
> Now, in this situation you disable IDE controllers. But, despite that, I
> do read in places, about telling BIOS about the drive parameters.
That would be if you were using an AT controller card.
Ignore that - your XT controller oncard-ROM is programmed with the drive
parameters internally.
--
Adrian C
DerekW
Mike Tomlinson
<ju...@jaj22.org.uk> writes
I'm a bit fuzzy about that too. Would have thought it would disable the
onboard BIOS so the controller could be used with the built-in AT setup
utility, but the OP seems fairly sure the XT BIOS is still there.
Personally, I'd set it to XT mode, get it going, then experiment with AT
mode.
--
Mike Tomlinson
Mike Tomlinson
writes
>Is it about right that it takes 5 hours to physically format the 20 meg
>drive?
No.
>Can you choose any value for cylinders, or do you have to go with 615 (the
>handbook value)?
Any value equal to or less than the handbook value. But for each
cylinder you reduce it by, you'll reduce the drive capacity by 0.034MB.
You could even format the drive with 20 cylinders if you wanted, but an
0.68Mb drive would be somewhat, um, less than useful.
612 is a common cylinder value for 20MB drives, which is why I suggested
it.
>Anyway, because the red LED lights up on the drive on (DOS) startup, most, I
>think, would have bet the drive is readable and writeable. But FDISK just
>won't show it's menu, because there is no drive accessibility, all FDISK
>will do is show there is a 20 meg disk installed, when the STATUS swith is
>used.
The drive isn't formatting properly, despite the Format Successful
message you get. It's either duff or you've input a value the
controller doesn't get along with.
--
Mike Tomlinson
Rich
"Mike Tomlinson" <mi...@none.invalid> wrote in message
news:V1B9a2F0...@none.invalid...
If the drive had to be initially low-level formatted (because it had never
been low-level formatted) at the moment I'm thinking the only way to do that
is to run DEBUG with -G=C800:5. I don't know of any other way myself. Well,
I'm doing that, I do the low-level formatting, but note that I have a jumper
set to "AT mode enabled". Otherwise it would be XT mode enabled.
I put it to AT mode, because I have an AT machine, and I thought unless I do
that, the controller certainly won't work. Not for any particular reason
such as BIOS or anything, I just thought that it was suppossed to be set the
AT mode. That's what I have been thinking. So, of course, I'm surprised at
you saying setset to XT mode. :c) You see, I'm not that up on PC's :c)
Interesting about, is this drive really getting formatted.
I've no idea what changing from XT to AT mode does. But like you say, does
AT mode divert the controller to use my regular BIOS. That probably is the
issue. But, if my card can be made an AT card, can I use my system BIOS, if
I disable IDE? Does my card become an IDE card when AT mode enabled?
Rich
"Adrian C" <em...@here.invalid> wrote in message
news:7lqbmjF...@mid.individual.net...
But is my card an AT card when it's set to AT mode enabled?
Rich
"Mike Tomlinson" <mi...@none.invalid> wrote in message
news:V1B9a2F0...@none.invalid...
Are we getting a conflict, because the controller card BIOS is set on,
whilst also set to AT mode enabled?
Rich
> On Mon, 9 Nov 2009 10:09:50 -0000, "Rich" <no...@emailo.com> wrote:
>
>>Is it about right that it takes 5 hours to physically format the 20 meg
>>drive? I mean, it is setting the disk geometry.
>
> No. From rusty memory, it should be between 10 minutes and half an
> hour.
>
> Excessive time would normally mean (a) that your interleave is too
> small - so rather than being able to jump from one cylinder to the
> next and catch the start of the new cylinder, it's missing and having
> to take an entire extra revolution before it gets going.
>
> Or (b) that the disk is damaged and going through many retries to seek
> or verify, but you'd be able to tell by the different clicky noises
> the dodgy bits would make.
I set cylinders to 50 and took about 3 0r 4 minutes.
BTW, when I run FDISK ar start up, then the LED on the HD blinks once, then
the message "Eror reading fixed drive".
So, it looks like FDISK is communicating with the HD. But, reporting error
reading.
Might be time to connect my other ST-225 and see what I get.
Mike Tomlinson
writes
>But is my card an AT card when it's set to AT mode enabled?
No, because to be an AT card, it would need to be 16-bit and to have the
necessary "fingers" to plug into the extension to the 8-bit slot on the
motherboard (i.e. it'll fill an entire ISA slot.)
Without the 16-bit AT extension, interrupts above 7 are not available,
and the first hard disk on an AT machine uses IRQ14. The second uses
IRQ15. You cannot access these from an 8-bit card.
I've already said it several times, but you seem to regard usenet as
write-only. Set your card to XT mode and work from there.
--
Mike Tomlinson
Rich
"Mike Tomlinson" <mi...@none.invalid> wrote in message
Wait a minute, do you know what I do here? How do you know I haven't done
that? How?
I don't report everything I do here. Gosh.
Rich
Jaimie Vandenbergh
>I don't report everything I do here. Gosh.
It makes troubleshooting a truckload easier if you give more clues,
though!
Cheers - Jaimie
--
"If you are allergic to a thing, it is best not to put that thing in
your mouth. Particularly if the thing is cats."
- Lemony Snicket, "The Wide Window"
bcoombes
Rich
Well, I'm tryin' my best. :c)
I've just asked Western Digital what this AT mode enabled feature was for. I
think folks think it was something to do with enabling regular system BIOS
to handle the HD /controller. But, who really knows, only WD probably.
I did make it so that the card was XT mode enabled and kept BIOS enabled.
But that made no difference.
In an effort to speed things up when messing around formatting, I've at the
moment set the HD to:
ccc h rrr ppp ee o
50 4 25 25 11 5
I just arbitrarily chose the 25's. Perhaps there are better figures for 50
cylinders. Interleave 6.
It is normally recommended that I use:
612 4 612 612 11 5
I've no idea about translation mode, LBA, PIO and all that good stuff you
can set in BIOS. But, so far, I know, try the XT mode. By that I think you
mean let the BIOS be the card BIOS. keep away for now from this AT mode
enable stuff.
-----
ccc = total number of cylinders (1-4 digits)
h = number of heads (1-2 digits)
rrr = starting reduced write current cylinder (1-4 digits)
ppp = write precomp cylinder (1-4 digits)
ee = max correctable error burst length (1-2 digits) range
= 5 to 11 bits, default = 11 bits
o = CBB option byte, step rate select (1 hex digit) range
Rich
"bcoombes" <bcoombes@orangedotnet> wrote in message
news:1e-dneMYfZ6ormXX...@brightview.co.uk...
Well, what I really meant is that I do take things onboard and then do some
messing. And I did change things to XT mode and ran the PC up.
Soon I'll probably recap steps that I've taken and if nothing else crops up
that can be done, I'll probably be quite for a while until and if I discover
something to enquire about.
I'm sort of cramming everything in time-wise so that's why all flurry of
posts recently. It's like I cannot put it down. :c)
DerekW
I opened the zip file it is there but probably not visable as it's a system
file
syntax > attrib -h command.com to unhide < the arcane part is of course
you need to sys the disk or it and the other two hidden system files will
not copy into the boot sector and I think format and fdisk are version
sensitive so you probably cannot use them with win versions.The boot disk
creator at http://www.bootdisk.com/bootdisk.htm might solve that then copy
across the other two files (format fdisk) on to your floppy
Derek
Rich
I said:
> In an effort to speed things up when messing around formatting, I've at
> the moment set the HD to:
>
> ccc h rrr ppp ee o
> 50 4 25 25 11 5
>
> I just arbitrarily chose the 25's. Perhaps there are better figures for 50
> cylinders. Interleave 6.
>
> It is normally recommended that I use:
>
> 612 4 612 612 11 5
>
> I've no idea about translation mode, LBA, PIO and all that good stuff you
> can set in BIOS. But, so far, I know, try the XT mode. By that I think you
> mean let the BIOS be the card BIOS. keep away for now from this AT mode
> enable stuff.
Woa! I'm getting ahead of myself here. I've not quite been getting it.
http://artofhacking.com/th99/c/U-Z/20217.htm
I ought to first try XT mode. Understood.
I have "SUPER BIOS".
Now, although the formatter enables me to format whatever cylinders I want
to, in dynamic format, in XT mode, there are only four HD options. And only
two of those options would match my 21Mb HD:
Cap Hds Cyls WP RWC
0 21mb 4 612 450 450
3 21mb 4 615 450 450
This means I don't have carte-blanche to pick & choose cylinders etc. I must
stick with what the table says.
Now, I suspect when it comes to AT mode-enabled this makes possible the use
of many more HD BIOS settings, but for that to happen BIOS must reside
elsewhere. I always have the necessary freedom in formatting, but not in the
BIOS options. That is probably what's going on with the AT mode thingy.
First thing, get the formatting right for XT mode!
I hope I'm getting it a bit more. :c)
Roger Hunt
>Rich wrote:
>> "Mike Tomlinson" <mi...@none.invalid> wrote in message
>> news:RUOsiNGK...@none.invalid...
>>> In article <7lqgs2F...@mid.individual.net>, Rich <no...@emailo.com>
>>> writes
>>>
>>>> But is my card an AT card when it's set to AT mode enabled?
>>>
>>> No, because to be an AT card, it would need to be 16-bit and to have the
>>> necessary "fingers" to plug into the extension to the 8-bit slot on the
>>> motherboard (i.e. it'll fill an entire ISA slot.)
>>>
>>> Without the 16-bit AT extension, interrupts above 7 are not available,
>>> and the first hard disk on an AT machine uses IRQ14. The second uses
>>> IRQ15. You cannot access these from an 8-bit card.
>>>
>>> I've already said it several times, but you seem to regard usenet as
>>> write-only. Set your card to XT mode and work from there.
>>
>> done that? How?
>> I don't report everything I do here. Gosh.
>>
>I'm beginning to worry about your skill-set Rich.
I'm agog at the tenacity!
--
Roger Hunt
bcoombes
again. You've got a F*****G XT card in a F*****G XT (8 bit) slot so you *have*
to use F*****G XT mode, forget AT...erase it totally from your memory banks.. :)
bcoombes
Yeah, it's awesome....or was that gruesome. :)
Roger Hunt
They *will* work, but perhaps not in the way the designers ever
envisaged.
--
Roger Hunt
Rich
"bcoombes" <bcoombes@orangedotnet> wrote in message
Hi. I don't think what is you say is actually true.
The 8 bit card has worked in AT machines:
http://www.vintage-computer.com/vcforum/showthread.php?t=17974
User: MODEM 7
The following should convince you of the ability of the WD1002A-WX1 to
use the ST-225:
I have a WD1002A-WX1 (dated 1987) with Super BIOS chip.
I placed the WD1002A-WX1 and an ST-225 in an XT clone (running at 4.77
MHz). Ensured that WD1002A-WX1 jumpered for XT mode. As expected, I had to
low-level format the ST-225 (because it had the format of a different
controller on it), which I did using the standard 'G=C800:5' method (and
then choosing Y at the dynamic configure question). Followed that with
FDISK/FORMAT to find that the ST-225 booted nicely.
I then grabbed one of my AT clones, a 486DX-66 machine (dated 1993). I
removed the existing HDD controller. I entered the CMOS SETUP of the machine
and configured 'no drives installed'. I then jumpered the WD1002A-WX1 for AT
mode then inserted it and the ST-225 into the machine. Machine successfully
booted from the ST-225.
I could try the same operation with one of my early Pentium machines.
The only difference would be that because the HDD controller is built into
the motherboard, I'd have to enter CMOS SETUP and disable the 'onboard IDE'
functionality there. I have no doubt that that would work."
User: DRUID6900
"It does. I took up where you left off with a WX1 card and 225 in a P1 233MX
MoBo. With the on-board HD channels disabled and no entries in the CMOS
drive table, the drive fdisked, partitioned, formatted, took DOS 5 and
booted."
Adrian C
>
> "bcoombes" <bcoombes@orangedotnet> wrote in message
>>>
>> say it again. You've got a F*****G XT card in a F*****G XT (8 bit)
>> slot so
>> you *have* to use F*****G XT mode, forget AT...erase it totally from your
>> memory banks.. :)
>
>
> Hi. I don't think what is you say is actually true.
>
> The 8 bit card has worked in AT machines:
We are not doubting that it doesn't. It's just that the card does not
communicate with the motherboard BIOS settings as it is physically not
an AT card, doesn't share the AT port settings, does not have the AT
interrupt lines fed to it.
It is XT and has it's own mind on drive type, CHS settings. The fact it
works on AT machines is called backwards compatibility. However it is
working in an XT mode, and completely ignores what has been entered in
the motherboard BIOS, it has it's own.
Reread what you have posted from the forum.
The XT/AT jumper does not determine mode as we can see. It probably may
have something else to do with the operation of the card. Dunno. XT/AT
Bus wait states?
Really keen to find out WD's response on that :-)
Can you try the other drive and use FDISK to show the partition information?
> http://www.vintage-computer.com/vcforum/showthread.php?t=17974
--
Adrian C
Mike Tomlinson
<bcoombes@orangedotnet.?> writes
>Since that Tomlinson guy is obviously trying to keep it a secret let me say it
>again. You've got a F*****G XT card in a F*****G XT (8 bit) slot so you *have*
>to use F*****G XT mode, forget AT...erase it totally from your memory banks.. :)
*cackle*
--
Mike Tomlinson
Rich
"Adrian C" <em...@here.invalid> wrote in message
All noted. Thanks. I know, I'm out of my depth. :c)
The question to WD is now "escalated" so I hope to get a reply.
I'm assuming what AT mode enabled might do, and probably getting it all
wrong.
When DOS loads up, it makes the onboard LED flicker about a second, so there
some kind of communication going on with the HD.
I (messing about) disabled BIOS on the card and had AT mode enabled
and set to TYPE 2 in sysytem BIOS, IDE disabled. (That's me thinking
AT mode simply means I can set HD parameters in my system BIOS).
FDISK reported no disk, so although one can disable BIOS on the
card, I think it must always be enabled. But,why you can disable
BIOS on the card, I haven't the foggiest.
bcoombes
God knows, we await WD's response with bated breath. :)
>>
>> Really keen to find out WD's response on that :-)
>>
>> Can you try the other drive and use FDISK to show the partition
>> information?
>>
>>
>
> All noted. Thanks. I know, I'm out of my depth. :c)
>
> The question to WD is now "escalated" so I hope to get a reply.
>
> I'm assuming what AT mode enabled might do, and probably getting it all
> wrong.
>
Yeah, it is a bit (very) confusing but as has been stated there is no way the
card can have AT (16 bit) functionality (jumper notwithstanding) as it is
physically in an 8 bit slot and can only read 8 bit signals. I'll be very
interested to hear what WD have to say as well.
Rich
"John Jordan" <ju...@jaj22.org.uk> wrote in message
news:m0KJm.4289$JI4...@newsfe21.ams2...
> Rich wrote:
>>
>> I'm running the old controller on "AT mode enabled". Because my system is
>> AT. But the WD1002A-WX1 is essentially an XT controller I guess.
>
> I'm still wondering what on earth this jumper does.
>
>
> John Jordan
Well, WD will not help, the reply was:
"We no longer have documentation for drives that old."
I'm not even sure the guy knows I'm talking about a contoller card rather
than a HDD.
So, no-one knows that AT mode enabled actually does.
bcoombes
One thing is for sure, it doesn't magically enable an 8 bit card to receive
signals from lanes it's not physically connected to so it can only be something
to do with the way the card communicates with the disk..as to what that is I
can't really hazard an even vaguely intelligent guess. :)
Roger Hunt
>Rich wrote:
>
>I'm still wondering what on earth this jumper does.
>
--
Roger Hunt
bcoombes
Feedlines R US :)
Rich
"John Jordan" <ju...@jaj22.org.uk> wrote in message
news:m0KJm.4289$JI4...@newsfe21.ams2...
> Rich wrote:
>>
>> I'm running the old controller on "AT mode enabled". Because my system is
>> AT. But the WD1002A-WX1 is essentially an XT controller I guess.
>
>
>
> John Jordan
What it probably does is switch the controller card from working in DMA
(channel 3) to PIO.
XT uses DMA 3 for HDD controller cards. But, when you have a faster AT
machine, it's better to use PIO.
I just cannot though get FDISK to write the MBR. It just cause the LED to
flicker once, then reports "Error reaading fixed drive". I could do with a
disk diagnostic.
Rich
"Jaimie Vandenbergh" <jai...@sometimes.sessile.org> wrote in message
news:kgbgf55vvt2934vt6...@4ax.com...
Okay, the card seems to be configurable for DMA3 and PIO.
First thing to do is to low-level format the drive with the
controller card. Now, there are no system BIOS settings needed to
enable this, the card itself simply accesses the HDD and formats it on:
DEBUG -G=C800:5. But, that is not getting the operating system
to communicate with the card/HDD. I format "Standard HD" that is a
615 cyl; 4 hds HDD. Interleave 3. That should be perfectly satisfactory
for my ST-225. I set the drive-type switches on the card to #3 which,
according to the cards drive-type table for SUPER BIOS is:cyl 615;hds 4;
WP 450;RWC 450. I think this is okay. In this scenario there is no
selecting of CHS or drive-type in system BIOS. Any BIOS settings is
done by jumpers on the disk controller card.
(One thing I do not understand. Per the handbook ST-225 figures are:
cyl 615;hds 4;WP 300; SPT 17. Choosing #3 on card is installing a drive
with slightly different values viz: cyl 615;hds 4;WP 450 ;RWC 450.
Probably makes no difference).
Anyway, I'm not dynamically formatting the drive as per the following
chart. I'm just letting the contoller do it using the default settings
in SUPER BIOS for a standard 10/20 Mb HDD. Maybe I should be dynamically
formatting - I don't know. I can change the interleave with the default
format or dynamic formatting. I suppose I could have a problem with
my formatting and not know it.
DYNAMIC FORMATTING CHART
ccc h rrr ppp ee o -where:
ccc = total number of cylinders (1-4 digits)
h = number of heads (1-2 digits)
rrr = starting reduced write current cylinder (1-4 digits)
ppp = write precomp cylinder (1-4 digits)
ee = max correctable error burst length (1-2 digits) range
= 5 to 11 bits, default = 11 bits
o = CBB option byte, step rate select (1 hex digit) range
= 0 to 7, default = 5 refer to controller and drive
specification for step rates.
DMA MODE
The card worked in an 8 bit ISA card. Under XT, data tranfers were made
via DMA3. Used IRQ5.
I have a 16 bit ISA card because I have an AT motherboard. DMA chip 1 is
DMA0-DMA3. DMA chip 2 is DMA4-DMA7. DMA1-DMA3 are for 8 bit transfers,
DMA5-DMA7 for 16 bit. I believe I must set DMA3 in my BIOS to ISA/EISA,
and IRQ5 to ISA/EISA for the card occupying the 8 bit section of the ISA
slot.
On the card itself I should select "XT mode enabled" which allows (one
thinks) the card to use DMA3.
I must disable IDE in BIOS, otherwise I think there will be a conflict
(Disable *certainly* I think in MAIN. See below.)
It could be wise to disable in BIOS any device using IRQ5 at this stage.
Now, I should run FDISK to lay standard MBR to the drive, then hi-level
format so that the OS can deal with the HDD. But FDISK reports: "Error
reading fixed drive", right after the LED on the HDD blinks.
Now in this situation, telling the PC about CHS or drive-type, is a
matter of setting jumpers on the card. There is no going into system
BIOS and choosing a HD type. Indeed IDE0-IDE3 are set "Not Installed". I
think this clears the relevant spots in the CMOS. Indeed, also Onboard
IDE I've disabled in BIOS PERIPHERAL SETUP.
BIOS BTW is AMI BIOS 1998.
Here are selected BIOS settings:
MAIN
Primary IDE Master: Not Installed
Primary IDE Slave: Not Installed
Secondary IDE Master: Not Installed
Secondary IDE Slave: Not Installed
(Of course Floppy installed)
ADVANCED CMOS SETUP
Display Mode At Add On ROM Init: Force BIOS
SMART For Hard Drives: Disabled
Event Logger: Enabled
Try Other Boot Devices: Yes
PCI PLUG & PLAY SETUP
Allocate IRQ To PCI VGA: No
DMA Channel 3: ISA/EISA
IRQ5: ISA/EISA
PERIPHERAL SETUP
Onboard FDC: Auto
Onboard Serial Port A: Disabled
Serial Port A: IRQ4 (Greyed)
Onboard Serial PortB: Disabled
Serial PortB: IRQ3 (Greyed)
IR i/o Pin Locatioin Select: IRRX/IRTX
Serial PortB Mode: IrDA
IR Duplex Mode@ N/A (Greyed)
IrDA Protocol: N/A (Greyed)
Onboard Parallel Port: Disabled
Parallel Port Mode: N/A (Greyed)
EPP Version: N/A (Greyed)
Parallel Port IRQ: N/A (Greyed)
Parallel Port DMA Channel: N/A (Greyed) NB: Usually Channel 3 on my PC.
Onboard IDE: Disabled
Hard Disk Delay Time: Disabled
So, that's it for DMA mode.
PIO MODE
In theory the card works in AT Mode. That means setting the jumper to AT
Mode Enabled, so tranfers are by Programmed Input / Output. Here the CPU
does the work of transfer rather than the DMA controllers. You do this
because with an AT machine, data tranfers will be quicker than DMA.
But, if you chose AT mode Enable, does that mean you have to set CHS or
drive-type in system BIOS? When a drive-type is selected in IDE Device
Confuration you see the following options:
LBA Mode: On or Off
Block Transfers: On or Off
Fast Programme i/o Mode: Auto or 0, 1, 2, 3, 4, 5
32 Bit Transfer Mode: On or Off.
I note that:
User: MODEM 7
"I have a WD1002A-WX1 (dated 1987) with Super BIOS chip.
I placed the WD1002A-WX1 and an ST-225 in an XT clone (running at 4.77
MHz). Ensured that WD1002A-WX1 jumpered for XT mode. As expected, I had to
low-level format the ST-225 (because it had the format of a different
controller on it), which I did using the standard 'G=C800:5' method (and
then choosing Y at the dynamic configure question). Followed that with
FDISK/FORMAT to find that the ST-225 booted nicely.
I then grabbed one of my AT clones, a 486DX-66 machine (dated 1993). I
removed the existing HDD controller. I entered the CMOS SETUP of the machine
and configured 'no drives installed'. I then jumpered the WD1002A-WX1 for AT
mode then inserted it and the ST-225 into the machine. Machine successfully
booted from the ST-225.
I could try the same operation with one of my early Pentium machines.
The only difference would be that because the HDD controller is built into
the motherboard, I'd have to enter CMOS SETUP and disable the 'onboard IDE'
functionality there. I have no doubt that that would work."
User: DRUID6900
"It does. I took up where you left off with a WX1 card and 225 in a P1 233MX
MoBo. With the on-board HD channels disabled and no entries in the CMOS
drive table, the drive fdisked, partitioned, formatted, took DOS 5 and
booted."
Here with DRUID6900, AT Mode Enabled is selected, but there is no setting
of CHS or drive-type in BIOS - I don't think.
Although sometimes I think I've seen folks mention select Type 2 for the
drive in BIOS. I don't know if this spurious or not. I'm confused about
what to do with system BIOS when AT Mode (PIO Mode)is selected.
What gets me is, does disabling both IDE in BIOS PERIPHERAL SETTINGS make
HD settings in MAIN (say: Primary IDE Master: Type 2)irrelevant or not.
I'm not entirely sure. I think BIOS never-the-less tries to install Type
2 HDD when there is a HDD type there.
Anyway, I guess if AT Mode is selected, one must reset DMA3 to PnP. I
guess, to prevent a possible conflict with PIO mode of tranfer. But,
keep IRQ5set to ISA/EISA, not PnP (Pnp: I think effectively PCI).
I should cut to the chase and try to get a HD diagnose program,
so what that says about the HD. Probably will access the HDD if the
drive is basically OK.
Thanks for everyone that made input. I think I've gone as far as I can with
this 8 bit card short of running a HD diagnose test.
Of course, there could be errors I'm making, particularly in my settings
selections, but I've tried several.
http://artofhacking.com/th99/h/txt/4017.txt
http://www.dcllabs.net/docs/1002awx1.txt
http://artofhacking.com/th99/c/U-Z/20217.htm
http://wiki.osdev.org/DMA
http://www.alasir.com/books/hards/080-082.html
http://artofhacking.com/th99/m/M-O/35986.htm
Mike Tomlinson
writes
>Thanks for everyone that made input. I think I've gone as far as I can with
>this 8 bit card short of running a HD diagnose test.
Agreed. Look for a copy of an old DOS diagnostic called CheckIt. It
has a hard drive diagnostic.
>Of course, there could be errors I'm making, particularly in my settings
>selections, but I've tried several.
I don't think so, all looks right (at last). I expect you just have a
duff drive(s). You haven't responded to my questions about the noise
the drives make.
--
Mike Tomlinson
Rich
"Mike Tomlinson" <mi...@none.invalid> wrote in message
Hi. When the drive starts up, it goes a few seconds, then it makes 2 beeps.
It takes about 5 hours to format. I think that's normal actually.
It blinks once just before FDISK reports error.
When Partition Magic started up in DOS mode, during the 10 seconds it took
to start, the red LED on the HDD was flashing away. Heard nothing though.
Partition Magic cannot apply any commands.
As to what I'm sure is PIO mode (AT Mode Enable): Because I saw to PIO in
system BIOS, I did wonder whether system BIOS had to roped-in in this mode.
But, I don't think so.
I can put a dump from System Analser (Product: SA53W) on a webpage. But, I
don't think that will assist.
I wonder if the PC is trying to work with SCSI.
I wonder if my BIOS is too "modern".
Adrian C
> What it probably does is switch the controller card from working in DMA
> (channel 3) to PIO.
>
Any refs Rich for where you found that? Not that we don't believe you -
but I (and no doubt a few others) must have been up and down google
looking at this kind of stuff.
Be nice to *know* where this info is if in the future I ever get
transported up on to some 'Antiques Roadshow' 3D-immersive hologram
sensurround III programme to identify some pre-flying saucer stuff.
> XT uses DMA 3 for HDD controller cards. But, when you have a faster AT
> machine, it's better to use PIO.
>
--
Adrian C
Roger Hunt
>Rich wrote:
>
>> What it probably does is switch the controller card from working in DMA
>> (channel 3) to PIO.
>
>Any refs Rich for where you found that? Not that we don't believe you -
>but I (and no doubt a few others) must have been up and down google
>looking at this kind of stuff.
>
>Be nice to *know* where this info is if in the future I ever get
>transported up on to some 'Antiques Roadshow' 3D-immersive hologram
>sensurround III programme to identify some pre-flying saucer stuff.
>
--
Roger Hunt
Rich
"Adrian C" <em...@here.invalid> wrote in message
Er, Chuck (G) said so. :c)
http://www.vintage-computer.com/vcforum/showthread.php?t=18089
But, could it be setting translation or not. Some early WD controllers you
can have translation or not. Chuck could be mistaken perhaps?
http://www.wdc.com/en/library/legacy/controllerboards/1003rah.pdf
But I reckon part of my controller is dead.
IC's make heat, even idle. When I put my face (chin actually) to IC U13,
(WD11C00) it never shows warmth like the rest. So, I reckon that IC is duff.
and that is the only thing that is stopping my card from working.
Rich
>
> But I reckon part of my controller is dead.
>
> IC's make heat, even idle. When I put my face (chin actually) to IC U13,
> (WD11C00) it never shows warmth like the rest. So, I reckon that IC is
> duff. and that is the only thing that is stopping my card from working.
But, what is confusing, is how can my card format, yet WD11C00 be dead. I'd
bet my money it's dead, but the card formats the drive. That to me is
weird.
nobody >
Without knowing the guts of that "U13", your "temp test" is invalid.
(and yes, I've troubleshot stuff the same way.)
Mike Tomlinson
writes
>Hi. When the drive starts up, it goes a few seconds, then it makes 2 beeps.
Normal.
>It takes about 5 hours to format. I think that's normal actually.
Wrong. Not normal.
As you have been asked previously, more than once: does the drive make
any noise during format? If it makes a gentle tick-tick-tick like a
clock running fast (and you may be able to see the screw in the stepper
motor turning v-e-r-y slowly), it's probably working. If it makes no
noise at all, it's fucked. Probably stiction.
>It blinks once just before FDISK reports error.
Normal.
>When Partition Magic started up in DOS mode, during the 10 seconds it took
>to start, the red LED on the HDD was flashing away. Heard nothing though.
>Partition Magic cannot apply any commands.
Suggests the drive is fucked.
>As to what I'm sure is PIO mode (AT Mode Enable): Because I saw to PIO in
>system BIOS, I did wonder whether system BIOS had to roped-in in this mode.
>But, I don't think so.
Nah. Forget about all that. (Please? pretty please?)
>I can put a dump from System Analser (Product: SA53W) on a webpage. But, I
>don't think that will assist.
No.
>I wonder if the PC is trying to work with SCSI.
No. "Boot from SCSI" is just another way of saying "Boot from a plug-in
card with its own BIOS". The word SCSI is used because in the era
during which your motherboard was made, it was fairly common for power-
users to plug in SCSI controllers and SCSI drives. All it does is cause
the system BIOS to pass control to an external controller at boot. It's
then up to that controller to get the boot record from the drive and
execute the boot code. That is clearly happening in your case.
>I wonder if my BIOS is too "modern".
No.
(where did I mine this rich seam of patience? I must be getting old.)
--
Mike Tomlinson
bcoombes
>
> (where did I mine this rich seam of patience? I must be getting old.)
>
Lol, hankering for the good old days maybe. There was something about those
Seagate MFM drives, they made a particularly pleasing sound during read/seek
ops, maybe that's why this thread generated so much interest. Anyway it
certainly sounds as if the drive has had it which comes as no surprise
really...it's quite a long way past it's sell by date.
Rich
"Mike Tomlinson" <mi...@none.invalid> wrote in message
> As you have been asked previously, more than once: does the drive make
> any noise during format? If it makes a gentle tick-tick-tick like a
> clock running fast (and you may be able to see the screw in the stepper
> motor turning v-e-r-y slowly), it's probably working. If it makes no
> noise at all, it's fucked. Probably stiction.
I'm sure I mentioned this, but just for you: :c)
First, the drive, in power up, makes the regular noises, first comes a short
sound like something is engaging, then 2 beeps. Sounded like that always.
As to formatting: There is regular sequence of trrrrrrrrrrbeep,
trrrrrrrrrrrrbeep, ....for about 5 hours. And the red LED is on all of the
time until formatting stops. These days I set interleave to 6 which might
make things better.
Rich
"nobody >" <useneth...@aol.com> wrote in message
news:t5idnSNaIcR9iWDX...@supernews.com...
http://www.datasheetarchive.com/WD11C00-17-datasheet.html
This is basically the IC. If this was broken or partly, I bet I'd be able
to format with DEBUG command: -G=8000:5 non-the-less.
Mike Tomlinson
writes
>I'm sure I mentioned this, but just for you: :c)
Ooh, thanks :)
>First, the drive, in power up, makes the regular noises, first comes a short
>sound like something is engaging, then 2 beeps. Sounded like that always.
That sounds ok.
>As to formatting: There is regular sequence of trrrrrrrrrrbeep,
>trrrrrrrrrrrrbeep, ....for about 5 hours. And the red LED is on all of the
>time until formatting stops.
That doesn't. All you should hear (apart from the drive spinning)
during a format is a gentle tick-tick-tick as the stepper motor moves
the heads to the next cylinder. The trrr-beep you describe is the drive
losing its bearings and trying to re-calibrate by moving the heads to
the last cylinder (612, 615, I won't go there again) and back to the
cylinder it is trying to format.
Head positioning on modern drives is done by a voice coil and each
platter has servo information embedded in the data tracks so the drive
can keep track of where its' heads are. On drives of the ST225 era, the
heads are moved by a stepper motor and formatting is done "blind". Each
track is formatted then read to ensure it is usable. If the read fails,
the head is recalibrated by moving it to the last cylinder and back.
When it makes this noise, you'll see the screw in the stepper motor wind
right round and whizz back again (be very careful moving a spinning
drive. Try and position it on its side in such a way it can't fall
over.)
It's possible that in your case one head has failed and therefore the
drive is unable to format. Usually head failure is terminal - the head
flies off its arm with subsequent damage to the platter surface
(scratches) - on that ST225 drive, when a head comes off you will see
radial scratches where the copper-ish colour of the platter is scratched
through to the aluminium surface. Lovely example at:
http://www-uxsup.csx.cam.ac.uk/~fanf2/hermes/doc/talks/2006-01-techlinks
/head_crash.jpg
If it's a failure of the head electronics (amplifier chip failures were
common), it just won't be able to read or write anything.
When you format again, try setting heads to 1 instead of 4. If that
works, try 2, then 3. When it stops working, you've found the failed
head, unless it won't even format with one head, then you know it has to
be head 1 or some other problem.
Finally, don't open the drive (the top is only held on by 4 screws IIRC)
unless you are prepared to chuck it in the bin. They're assembled in
clean rooms and even a speck of dust can cause a fatal head crash.
--
Mike Tomlinson
Rich
"Mike Tomlinson" <mi...@none.invalid> wrote in message
Here are some test results of a regular format:
I've marked the spindle on the stepper so I can observe it's motion.
On PC power-up: from it's resting or start or original position,
the spindle turns very quickly a few times one way, then returns
to the start position. Then 2 beeps.
Format: When Formatting begins the spindle hardly moves at first,
but soon you can see it's moving to a new position very fast, then
HD brrrrrs whilst the spindle makes it's way back the starting
position. Then, once at the original position again, it again
quickly flicks away to the new postion, slowly make it's way
back the start once again, all the while HD brrring. The brrring
only takes place as the spindle is making it's way back to start.
As time goes by the new position becomes further and further away from
start.
After about 1 hour it flicks 1/2 of a turn, then brrrrrrs all the way
to original position. So, after 5 hours it must then, if things continue,
flick about 2 1/2 times before returning to start. So, the brrrring gets
longer and longer.
I Ctrl + Alt + Delete, which left the spindle away from it's original
position. When PC booted itself, the HD onboard LED flickered, and the
spindle slowly reposioned to original position.
Is the normal for an old ST-225 or not?
Jaimie Vandenbergh
>Is the normal for an old ST-225 or not?
Not. It's resetting the heads after every track read failure. Sounds
like every single track read is failing, so what you have there is
just an entertaining doorstop.
Cheers - Jaimie
--
If you can't behave right, then don't behave at all.
Mike Tomlinson
writes
>Here are some test results of a regular format:
>
>I've marked the spindle on the stepper so I can observe it's motion.
Yay! I was going to suggest this :) Most St225 spindles have a
"screwdriver" slot in their base which makes movement very visible, but
some don't. A dab of Tippex helps on those.
>On PC power-up: from it's resting or start or original position,
>the spindle turns very quickly a few times one way, then returns
>to the start position. Then 2 beeps.
Sounds ok.
>Format: When Formatting begins the spindle hardly moves at first,
>but soon you can see it's moving to a new position very fast,
i.e. winds to cyl 612 (trrrrrrrrrrrr)
> then
>HD brrrrrs whilst the spindle makes it's way back the starting
>position.
(beeeeeeeeeeep)
> Then, once at the original position again, it again
>quickly flicks away to the new postion,
The controller has given up on the unusable track and told the drive to
try and start formatting the next track along.
> slowly make it's way
>back the start once again, all the while HD brrring.
the format has failed, so it's trying to recal by seeking to cyl 625 and
back
> The brrring
>only takes place as the spindle is making it's way back to start.
Yes, good.
>As time goes by the new position becomes further and further away from
>start.
As you would expect - it's trying to format subsequent tracks.
>I Ctrl + Alt + Delete, which left the spindle away from it's original
>position. When PC booted itself, the HD onboard LED flickered, and the
>spindle slowly reposioned to original position.
that's normal. Ctrl-Alt-Del sends a reset signal to the hardware, so
the controller responds by resetting everything to a nominal state.
>Is the normal for an old ST-225 or not?
Nope.
Can you beg/steal/borrow another drive, preferably known working? Any
20Mb MFM drive will do. Look for Miniscribe 8425, Kalok KL320, Western
Digital WD262, etc. These were all popular drives of the era.
--
(\__/)
(='.'=) Bunny says Windows 7 is Vi$ta reloaded.
(")_(") http://imgs.xkcd.com/comics/windows_7.png
bcoombes
> On Fri, 13 Nov 2009 18:15:41 -0000, "Rich" <no...@emailo.com> wrote:
>
>> Is the normal for an old ST-225 or not?
>
> Not. It's resetting the heads after every track read failure. Sounds
> like every single track read is failing, so what you have there is
> just an entertaining doorstop.
>
Even if he can't get it working it will easily take the title of 'Most discussed
drive on usenet' :)
Rich
"Mike Tomlinson" <mi...@jasper.org.uk> wrote in message
news:y7sPaQBmtd$KF...@jasper.org.uk...[....]
> Can you beg/steal/borrow another drive, preferably known working? Any
> 20Mb MFM drive will do. Look for Miniscribe 8425, Kalok KL320, Western
> Digital WD262, etc. These were all popular drives of the era.
Well, I have second ST-225, but I think it's doing the same. Ill mess.
Actually it's really a beep sound first, then brrrr. :c)
When the spindle flicks (turns quickly) from it's original starting position
to some other position, that is making a beep sort of sound. So, the quick
turn is first. Then, when it makes it's way back to start position from
wherever it got to, it's brrrrring all the way back. As I say, first these
sharp turns on the shaft start out tiny, but by 1 hour these turns have got
to 180 degrees from start, and brrrrring (slowly) all the way back to start.
So, the noise is an effort to recalibrate.
So, you are saying start position is cyl. 512. Therefore, it's formatting
cylinders backwards, i.e, 512, 511, 510,.... If I chose 50 clinders that
start position would be on cyl. 50 and part way down the platter would be
cyl.0?
Normal formatting then, the spindle shows a steady turning progression, not
going back to start all the time. I'll have to look out for that. :c)
One thing, if the disk always returns to start after it's reached a
cylinder, no part of the disk is readable it is assumed. Mmm, that's very
unusual. But, whether it's the ouboard controller, or HD internal circuitry
thats driving the stepper in it's programmed fashion, the stepper does move
in this manner when the controller cards formatting program is envoked
(G=C800:5). So, somethings working.
If all the disk was unformattable, the HD should still take a seek command
from some low-level HD diagnose program. I should see if I can position the
heads.
I think it's not the platter, but the electronics. :c)
Johnny B Good
from "Rich" <no...@emailo.com> contains these words:
Have you tried swapping the data cable to the other header on the
controller to see if it makes any difference?
--
Regards, John.
Please remove the "ohggcyht" before replying.
The address has been munged to reject Spam-bots.
Mike Tomlinson
writes
>Well, I have second ST-225, but I think it's doing the same. Ill mess.
Hm.
>So, you are saying start position is cyl. 512.
No, it's usually cyl 0 (the outermost one.)
> Therefore, it's formatting
>cylinders backwards, i.e, 512, 511, 510,....
I'd be surprised, but prepared to believe it.
> If I chose 50 clinders that
>start position would be on cyl. 50 and part way down the platter would be
>cyl.0?
No, I would expect it to start at 0 (the outermost) and work in 50
cylinders.
>Normal formatting then, the spindle shows a steady turning progression, not
>going back to start all the time.
Yes.
>One thing, if the disk always returns to start after it's reached a
>cylinder, no part of the disk is readable it is assumed. Mmm, that's very
>unusual.
Yes. I'm starting to wonder if you have the cabling right.
> But, whether it's the ouboard controller, or HD internal circuitry
>thats driving the stepper in it's programmed fashion
The controller send a -STEP signal to the drive telling it to move the
heads. 50 pulses of -STEP = cylinder 50.
>, the stepper does move
>in this manner when the controller cards formatting program is envoked
>(G=C800:5). So, somethings working.
It all sounds very close to working, I think you may have been unlucky
with the drives. Hence my suggestion to borrow another.
>If all the disk was unformattable, the HD should still take a seek command
>from some low-level HD diagnose program.
Maybe. The diagnostic may still expect to be able to read data.
Mike Tomlinson
<jcs.comp...@plugzetnet.co.uk> writes
> Have you tried swapping the data cable to the other header on the
>controller to see if it makes any difference?
I think that'll just make the drive play dead.
But was wondering if the OP's got the control cable (the 34 way one)
right. He seemed unsure about the twist and setting the drive ID.
Rich
After I posted, I knew I'd got things wrong.
Of course, start position is cyl 0. The heads flick to the cylinder to be
formatted, then for some reason, "brrrr" all the way back to start. Which
you say is some kind of recalibration. And that is ocurring for every
cylinder that's attempted to be formatted. So, it takes 5 hours. It's a
strange fault, but at the end, the format program never returns a fault. I
think perhaps there is no write or read taking place, and that's why the
entire drive looks bad. It's a write read problem. I muse.