> > like this one:http://www.tigerdirect.ca/applications/SearchTools/item-details.asp?E.... Mine may be a slightly older model but it looks a LOT like this one. There's is a red LED (or maybe two) just above the fan and it is on solidly whenever the computer is powered on. I don't recall that coming on before but I may not have noticed. If it isn't normally on, it may be providing a big clue about what my problem is. Unfortunately, I'm not positive if it is always on and I can't find anything in its manual that indicates the purpose of that LED. So maybe this is not relevant to my problem after all. This is a link to the manual:http://static.highspeedbackbone.net/pdf/Zerotherm_BTF_Series_Cooler_M...

>
> > My desktop is completely useless in its current state and I really need
> > to get it working again. I'm not sure what I've done or why removing the
> > new memory didn't eliminate the problem. Any help would be very much
> > appreciated! I'm just glad I got a laptop for Christmas or I'd be having
> > a much harder time posting this question....
>
> The manual mentions LEDs, but not their function.
>

> http://static.highspeedbackbone.net/pdf/Zerotherm_BTF_Series_Cooler_M...

>
> They appear to be simple illumination, and located
> in the fan body. They really should not flash, if the
> power feeding the fan is OK. If the actual LED is flashing,
> then you'd expect plenty of stuff on the computer to appear
> to be going on and off. That would be a power supply problem,
> or a problem with PS_ON# signal on the main connector cable.
>
> http://www.anandtech.com/show/2299/2
>

Okay, then the BTF90 LED was probably a red herring (as well as being
a red LED). I just never noticed that LED before.

> *******
>
> 1) Check that the monitor cable is seated.
>

Yes, it definitely is. On both ends.

> 2) If you're using a video card, make sure the screw
> holding the video cardfaceplate, isn't loose, and the video card
> is half out of the slot. That would prevent the monitor
> from working. Modern video cards have a "heel clamp", to
> prevent that from happening.
>

I'm not seeing the screw but the card looks properly seated to me. I
can take photos if you like and put them somewhere you can see them.

> 3) If you removed the ATX12V 2x2 square connector, that will
> stop the computer from posting.
>

I'm not sure where that is and haven't found it in the manual. But I
didn't do anything but shut the computer down before inserting the
memory. I shut it down again before removing the new memory.

> 4) Continue doing visual inspection, to see if you removed
> anything.
>

See my answer to point 3.

> 5) Remove all RAM, with all power off. *Never* insert or remove
> RAM, while the system is powered. On an Asus motherboard,
> when the green LED is not lit, only then is it safe to
> change memory. The green LED monitors +5VSB, and when the
> LED is extinguished, then there is no standby power in the RAM
> slots.
>

Uh oh. I didn't know the green LED had to be off. I did not turn it
off (via the power switch on the back or disconnecting the power for
the case) before inserting or removing the memory. That's never been
necessary on previous computers. Or maybe it was and I just got
lucky.

> 6) With all RAM removed, switch on. The case speaker connected to
> SPKR on the panel header, that speaker should start beeping. The
> beeping tells you:
>
> a) The processor is powered (ATX12V 2x2 is in place).
> b) The case speaker is connected.
> c) The processor ran some BIOS code.
> d) The processor programmed the speaker beeping (two or three
> beeps, and a repetitive pattern). The ability to continue the
> beep pattern, is supervised by CPU code. So much of the computer
> is running, if you heat the beeps. The computer is "running the
> beep code" and has not crashed. A completely silent computer
> (no beeps) is in serious trouble.
>

I turned off the LED and removed all the memory. No beeps at all when
I powered up.

What's a panel header and where is it? I'm not sure how to tell if the
SPKR is connected to it.

Have I hosed the whole computer here - motherboard, CPU and memory? Or
is the damage limited to (possibly) all of the memory, both old and
new?

--
Rhino

Paul

unread,

Apr 18, 2013, 2:22:59 PM4/18/13

to

hugh....@hotmail.com wrote:

>> 5) Remove all RAM, with all power off. *Never* insert or remove
>> RAM, while the system is powered. On an Asus motherboard,
>> when the green LED is not lit, only then is it safe to
>> change memory. The green LED monitors +5VSB, and when the
>> LED is extinguished, then there is no standby power in the RAM
>> slots.
>>
> Uh oh. I didn't know the green LED had to be off. I did not turn it
> off (via the power switch on the back or disconnecting the power for
> the case) before inserting or removing the memory. That's never been
> necessary on previous computers. Or maybe it was and I just got
> lucky.

Uh oh.

>
>> 6) With all RAM removed, switch on. The case speaker connected to
>> SPKR on the panel header, that speaker should start beeping. The
>> beeping tells you:
>>
>> a) The processor is powered (ATX12V 2x2 is in place).
>> b) The case speaker is connected.
>> c) The processor ran some BIOS code.
>> d) The processor programmed the speaker beeping (two or three
>> beeps, and a repetitive pattern). The ability to continue the
>> beep pattern, is supervised by CPU code. So much of the computer
>> is running, if you heat the beeps. The computer is "running the
>> beep code" and has not crashed. A completely silent computer
>> (no beeps) is in serious trouble.
>>
> I turned off the LED and removed all the memory. No beeps at all when
> I powered up.

Uh oh.

On the Asus M3A, the CPU is AMD, and the memory
controller is on the processor. Your first step, would
be to replace the CPU. The motherboard should be OK.

This will require removing the cooler (without snapping
off a CPU socket plastic tab), lift lever, remove CPU,
put in known working CPU, close lever, apply thermal paste
to top of CPU, clean and reseat CPU cooler, fasten clamp to tabs.
And so on.

Valid CPUs to use, are listed here. This is for an M3A.

http://support.asus.com/Cpusupport/List.aspx?SLanguage=en&m=M3A&p=1&s=24

And I'm assuming there isn't some other component which is
disconnected, not seated properly, bumped, etc.

>>
> What's a panel header and where is it? I'm not sure how to tell if the
> SPKR is connected to it.

The PANEL header is just below the green "SB_PWR" LED.

http://imageshack.us/scaled/landing/132/m3a.gif

>
> Have I hosed the whole computer here - motherboard, CPU and memory? Or
> is the damage limited to (possibly) all of the memory, both old and
> new?
>
> --
> Rhino

As you're not getting a beep from the speaker, I'd guess CPU.
The memory controller is on the CPU.

The same symptoms would result, if the ATX12V 2x2 is not connected.
That connector is square, and in the upper left of the m3a.gif above.
The m3a.gif is just copied from a PDF version of your Asus motherboard
manual. And this picture, shows another example of what to look for.
2x2 connector.

http://www.playtool.com/pages/psuconnectors/12v4pin.jpg

Important to note at this time, is whether you normally hear a
single "beep" when the computer was working. That single beep
is there, to prove the speaker works, at each startup. If they
didn't need to test the speaker, there would never have been
a single beep necessary, for the working state of the computer.
But the beep helps assure us the beep speaker is present and
accounted for. Then, on occasions there are no beeps of any
kind, it's easier to predict a catastrophic hardware failure
(power supply bad, power supply disconnected, CPU bad, half-seated
hardware, corrupted BIOS flash chip, flat CMOS battery, etc).

The PANEL header in the lower right hand corner, should have
wires to it like this. The thing is 2x10 form factor (room for
20 pins, not all pins installed). The SPKR connector is a 1x4
form factor pin header style connector, with wires installed
in the pin 1 and pin 4 positions.

PLED SPKR
| | | |
. . . . . . . . . .
. . . . . . . . . .
| | | | | |
IDE PWR RST
LED Switch Switch

You could take the tower to a computer store (repair facility),
and see if they have an AMD AM2 from the supported CPU list,
to test with. Or, they could test your CPU, in their motherboard.
That way, you'd be increasing the odds of making a correct diagnosis.
Or, you can take your chances on some Ebay processor, do the
prep yourself, and see what happens.

HTH,
Paul

hugh....@hotmail.com

unread,

Apr 18, 2013, 4:38:06 PM4/18/13

to

On 18 Apr, 14:22, Paul <nos...@needed.com> wrote:

So I can choose anything from that list, any of the Athlon/Phenom/
Semprons?

>
> And I'm assuming there isn't some other component which is
> disconnected, not seated properly, bumped, etc.
>

I did touch the video card as I was messing with the memory but it
looks properly seated. I also hooked up the laptop to the desktop
monitor (via its analog port) and it is working fine so I think that
removes any reasonable doubt about the monitor, if not the video
card.

>
>
> > What's a panel header and where is it? I'm not sure how to tell if the
> > SPKR is connected to it.
>
> The PANEL header is just below the green "SB_PWR" LED.
>
> http://imageshack.us/scaled/landing/132/m3a.gif
>
>
>
> > Have I hosed the whole computer here - motherboard, CPU and memory? Or
> > is the damage limited to (possibly) all of the memory, both old and
> > new?
>
> > --
> > Rhino
>
> As you're not getting a beep from the speaker, I'd guess CPU.
> The memory controller is on the CPU.
>
> The same symptoms would result, if the ATX12V 2x2 is not connected.
> That connector is square, and in the upper left of the m3a.gif above.
> The m3a.gif is just copied from a PDF version of your Asus motherboard
> manual. And this picture, shows another example of what to look for.
> 2x2 connector.
>
> http://www.playtool.com/pages/psuconnectors/12v4pin.jpg
>

I finally found the ATX 12V. I have the same picture in my manual and
just missed it.

> Important to note at this time, is whether you normally hear a
> single "beep" when the computer was working. That single beep
> is there, to prove the speaker works, at each startup. If they
> didn't need to test the speaker, there would never have been
> a single beep necessary, for the working state of the computer.
> But the beep helps assure us the beep speaker is present and
> accounted for. Then, on occasions there are no beeps of any
> kind, it's easier to predict a catastrophic hardware failure
> (power supply bad, power supply disconnected, CPU bad, half-seated
> hardware, corrupted BIOS flash chip, flat CMOS battery, etc).
>
> The PANEL header in the lower right hand corner, should have
> wires to it like this. The thing is 2x10 form factor (room for
> 20 pins, not all pins installed). The SPKR connector is a 1x4
> form factor pin header style connector, with wires installed
> in the pin 1 and pin 4 positions.
>
> PLED SPKR
> | | | |
> . . . . . . . . . .
> . . . . . . . . . .
> | | | | | |
> IDE PWR RST
> LED Switch Switch
>

Thank you! Now I know where the header is ;-)

> You could take the tower to a computer store (repair facility),
> and see if they have an AMD AM2 from the supported CPU list,
> to test with. Or, they could test your CPU, in their motherboard.
> That way, you'd be increasing the odds of making a correct diagnosis.
> Or, you can take your chances on some Ebay processor, do the
> prep yourself, and see what happens.
>

I'm pretty sure the place where I bought the memory has a repair
service. They seem fairly knowledgeable, at least within their
individual areas of expertise. I'll bring the tower, manual and memory
down there once you've confirmed that I can choose absolutely anything
from that list of CPUs. It seems like such a long list so I just want
to be sure. Then we'll see if they can test my CPU and memory in
something they have. Hopefully, I won't be out to much money for my
ignorance about the green LED....

--
Rhino

Paul

unread,

Apr 18, 2013, 8:46:31 PM4/18/13

to

One problem is, the availability of those processors. Anything
new is AM3 now. I did a quick check, and didn't get a match.
I tried Newegg and NCIX (Canada).

I was hoping eventually, maybe you'd mention the existing
model of processor. So I can narrow down things to take its
place. For example, if your existing processor is 89 watts,
I'd look for something with the same power envelope. As we
know your machine worked with an 89W processor, so another
89W processor should be OK.

Your shop, might not even have an AM2. Or, if the shop is at
all clever, they use a Sempron, say something with a single
core, that is dirt cheap, for testing. And consequently, if
any parts are "left over", the processor they've got might be
rather gutless.

If you no longer have the processor box as a reference,
you could always look for the receipt from the purchase.

*******

In the Phenom section of the CPU list, there are processors
like 9600 and 9650. The first one has a slight bug, while
the ones ending in xx50, are patched. So that cuts out a few
of the model numbers.

Paul

hugh....@hotmail.com

Apr 20, 2013, 5:27:47 PM4/20/13

to

hugh....@hotmail.com wrote:

>
> Eureka! We have beeping!

Good work.

Paul

hugh....@hotmail.com

unread,

Apr 20, 2013, 8:43:57 PM4/20/13

to

And yet I used to get a bright clear beep from the computer before my
recent troubles. I'm baffled by how I got that even though nothing is
connected to those speaker pins in the header, unless perhaps the
external speaker was generating the same beeps.

> The portable speaker you've been sold, is likely a piezoelectric type.
> It's a round black disc, about the size of a quarter or a dollar coin,
> with two electrical connections on it. For computer usage, it will have
> a red/black twisted pair of wires connected. And, we would hope, have
> a 1x4 pin spacing. (Red wire on pin 1, nothing in 2, nothing in 3,
> black wire on pin 4.) That would mate to your M3A SPKR pins.
>

Actually, the tech didn't sell it to me, he simply gave it to me. I
think he may have meant it to be a loan though so I'll offer it back
to him in case that's what he meant. And it doesn't look quite like
what you're describing. It's more of a cylinder than a coin. (When
would you ever have seen a dollar coin? I'm assuming you're American
and I've never seen dollar coins on your side of the border ;-)

The piezo he gave me had a 1x4 connector and GR (Ghostrider) said the
orientation didn't matter so I connected it and it worked.

> a miracle. Here's ...
>
> read more »

Okay, I've done my testing. Frankly, I have no idea what the results
mean but let me describe what I did so that you can verify that I did
everything reasonably and in a way that doesn't make the results
meaningless.

First, let me describe the memory chips with a short descriptive name,
the model number of the chip (and size), and the position where it
should be located once everything is working.

Name Type Position
------ ---------------------------- ---------
Old1 KVR800D2N5K2 (1GB) Old - A1
Old2 KVR800D2N5K2 (1 GB) Old - A2
New1 KVR800D2N6K2 (1 GB) New - B1
New2 KVR800D2N6K2 (1 GB) New - B2

I was tasked to test each memory card in each of the 4 slots with all
of the remaining slots left empty. I did exactly that, then did a few
additional tests out of personal curiousity. In each case, I cut the
AC power to the tower via the toggle switch at the back of the case
and waited for the green LED on the mobo to go off. In each case, I
tested a given setup at least three times and, in a couple of cases, 5
consecutive times. The results were as follows. I hope this doesn't
come out as a dog's breakfast; I'm using Google Groups and there
doesn't seem to be any way to make it display my post in a fixed font
although I've tried.

MEMORY(S) SLOT(S) BEHAVIOUR
----------- -----------
------------------
Old1 A1 CPU fan turns
off - pause – beep – pause - double-beep
Old1 B1 No beeps (CPU
fan doesn't stop)
Old1 A2 CPU fan turns
off – pause- beep – pause – double-beep
Old1 B2 No beeps (CPU
fan doesn't stop)
Old2 A1 CPU fan turns
off – pause – beep – pause - double-beep
Old2 B1 No beeps (CPU
fan doesn't stop)
Old2 A2 CPU fan turns
off – pause – beep- pause – double-beep
Old2 B2 No beeps (CPU
fan doesn't stop)
New1 A1 CPU fan turns off
– pause – beep – pause - double-beep
New1 B1 No beeps (CPU fan
doesn't stop)
New1 A2 CPU fan stops –
pause – beep – pause – beep – pause – double-beep (first time)
CPU fan
stops – pause – beep – pause – double-beep (next four times)
New1 B2 No beeps (CPU
fan doesn't stop)
New2 A1 CPU fan stops –
pause – beep – pause - double-beep
New2 B1 No beeps (CPU
fan doesn't stop)
New2 A2 CPU fan stops –
pause – beep – pause – beep – pause – double-beep (first time)
CPU fan
stops – pause – beep – pause – double-beep (next four times)
New2 B2 No beeps (CPU
fan doesn't stop)
Old1/ Old2 A1/A2 CPU fan stops –
pause – beep – pause – beep
New1/New2 A1/A2 CPU fan stops – pause
– beep – pause – beep
New1/New2 B1/B2 No beeps (CPU fan
doesn't stop)
Old1/New1Old2/New2 A1/B1/A2/B2 No beeps (CPU fan doesn't stop)

To summarize this, there were only four different behaviours that
occurred:
Behaviour 1: the CPU fan stopped after a few seconds, then there was a
pause, a single beep, another pause and a double beep.
Behaviour 2: the CPU fan stopped after a few seconds, then there was a
pause, a single beep, another pause, a single beep, another pause,
then a double beep.
Behaviour 3: the CPU fan stopped after a few seconds, then there was a
pause, a single beep, another pause, then a single beep
Behaviour 4: In all other cases, the CPU fan did not stop and there
were no beeps.

Behaviour 1 was noted when there was a single memory card installed
and the slot used was A1 or A2 (regardless of using old or new memory
card)

Behaviour 2 was noted only the first time a new memory card (either
one) was used in slot A2; subsequent tests of the same configuration
produced Behaviour 1

Behaviour 3 was noted when: there were two memory cards installed and
the slots used were A1 and A2 (regardless of whether the cards were
old or new)

Behaviour 4 was noted in all other cases.

I don't know if the CPU fan shutting itself off proves anything or is
completely irrelevant. Given that the B1/B2 slots behaved differently
than the A1/A2 slots and that the A1/A2 slots have been used
successfully for several years, is it possible that the B1/B2 slots
aren't turned on? In other words, is it possible that the B1/B2 slots
aren't being recognized yet because they need to have a jumper set or
have something done in the BIOS (when we can get to the BIOS again) to
ensure that those slots are active?

Paul

unread,

Apr 20, 2013, 9:22:38 PM4/20/13

to

I snipped this section and moved it down below... In case
Google shortens the post later.

The AMI beep codes are listed here.

http://www.bioscentral.com/beepcodes/amibeep.htm

The "pause ï¿½ beep ï¿½ pause - double-beep" is listed as video failure.
Which is curious. The "A" channel seems to be working (as a video
test probably comes after RAM test). Using "B" channel at all,
seems to stop it dead (probably crashed in BIOS).

MEMORY SLOT BEHAVIOUR
------ ---- ------------------
Old1 A1 CPU fan turns off - pause ï¿½ beep ï¿½ pause - double-beep

Old1 B1 No beeps (CPU fan doesn't stop)

Old1 A2 CPU fan turns off ï¿½ pause - beep ï¿½ pause ï¿½ double-beep

Old1 B2 No beeps (CPU fan doesn't stop)

Old2 A1 CPU fan turns off ï¿½ pause ï¿½ beep ï¿½ pause - double-beep

Old2 B1 No beeps (CPU fan doesn't stop)

Old2 A2 CPU fan turns off ï¿½ pause ï¿½ beep - pause ï¿½ double-beep

Old2 B2 No beeps (CPU fan doesn't stop)

New1 A1 CPU fan turns off ï¿½ pause ï¿½ beep ï¿½ pause - double-beep

New1 B1 No beeps (CPU fan doesn't stop)

New1 A2 CPU fan stops ï¿½ pause beep pause beep pause double-beep
(first time)
CPU fan stops ï¿½ pause beep pause double-beep

(next four times)
New1 B2 No beeps (CPU fan doesn't stop)

New2 A1 CPU fan stops ï¿½ pause ï¿½ beep ï¿½ pause - double-beep

New2 B1 No beeps (CPU fan doesn't stop)

New2 A2 CPU fan stops ï¿½ pause beep pause beep pause double-beep
(first time)
CPU fan stops ï¿½ pause beep pause double-beep

(next four times)
New2 B2 No beeps (CPU fan doesn't stop)

Old1/Old2 A1/A2 CPU fan stops ï¿½ pause ï¿½ beep ï¿½ pause ï¿½ beep
New1/New2 A1/A2 CPU fan stops ï¿½ pause ï¿½ beep ï¿½ pause ï¿½ beep

New1/New2 B1/B2 No beeps (CPU fan doesn't stop)
Old1/New1Old2/New2 A1/B1/A2/B2 No beeps (CPU fan doesn't stop)

So now we get to review your video card situation. The AMD 770
has no integrated graphics, so now we're looking at the video card
and the PCI Express x16 slot. Does the card have an auxiliary
power connector ? Is it connected ? Make and model of video card ?

http://en.wikipedia.org/wiki/AMD_770#770

Paul

hugh....@hotmail.com

unread,

Apr 21, 2013, 5:43:35 PM4/21/13

to

On 20 Apr, 21:22, Paul <nos...@needed.com> wrote:

> The "pause – beep – pause - double-beep" is listed as video failure.

> Which is curious. The "A" channel seems to be working (as a video
> test probably comes after RAM test). Using "B" channel at all,
> seems to stop it dead (probably crashed in BIOS).
>

I'm not sure what "crashed in BIOS" means. Does that suggest that
there's not a hardware failure and reconfirming a setting in the BIOS
will fix that problem?

Why don't the other combinations (Behaviours 2 and 3) appear in the
beep code table?
Or is that not a full list of all the possible beep codes? I found
that chart too but with the other combinations missing, I was afraid I
had the wrong table or one that wasn't applicable for this model of
motherboard.

> MEMORY SLOT BEHAVIOUR
> ------ ---- ------------------

> Old1 A1 CPU fan turns off - pause – beep – pause - double-beep

> Old1 B1 No beeps (CPU fan doesn't stop)

> Old1 A2 CPU fan turns off – pause - beep – pause – double-beep

> Old1 B2 No beeps (CPU fan doesn't stop)

> Old2 A1 CPU fan turns off – pause – beep – pause - double-beep

> Old2 B1 No beeps (CPU fan doesn't stop)

> Old2 A2 CPU fan turns off – pause – beep - pause – double-beep

> Old2 B2 No beeps (CPU fan doesn't stop)

> New1 A1 CPU fan turns off – pause – beep – pause - double-beep

> New1 B1 No beeps (CPU fan doesn't stop)

> New1 A2 CPU fan stops – pause beep pause beep pause double-beep
> (first time)
> CPU fan stops – pause beep pause double-beep

> (next four times)
> New1 B2 No beeps (CPU fan doesn't stop)

> New2 A1 CPU fan stops – pause – beep – pause - double-beep

> New2 B1 No beeps (CPU fan doesn't stop)

> New2 A2 CPU fan stops – pause beep pause beep pause double-beep
> (first time)
> CPU fan stops – pause beep pause double-beep

> (next four times)
> New2 B2 No beeps (CPU fan doesn't stop)
>

> Old1/Old2 A1/A2 CPU fan stops – pause – beep – pause – beep
> New1/New2 A1/A2 CPU fan stops – pause – beep – pause – beep

>
> New1/New2 B1/B2 No beeps (CPU fan doesn't stop)
> Old1/New1Old2/New2 A1/B1/A2/B2 No beeps (CPU fan doesn't stop)
>
> So now we get to review your video card situation. The AMD 770
> has no integrated graphics, so now we're looking at the video card
> and the PCI Express x16 slot. Does the card have an auxiliary
> power connector ? Is it connected ? Make and model of video card ?
>
> http://en.wikipedia.org/wiki/AMD_770#770
>
> Paul

The video card is an NVIDIA GEFORCE 9500GT according to the receipts.
It's interesting that you mention the video card. There is one small
thing I noticed when I was doing my tests. I was surprised to see a
rather large fan mounted to the bottom of the card. I may have seen
that before at some point but I'd completely forgotten that it had a
fan of its own, over and above the CPU fan and the case fans. Anyway,
just out of curiousity, I tried spinning the fan with my finger and
found that it didn't turn. I had assumed it would freewheel when not
being driven by electricity but it didn't budge. I didn't apply any
real pressure for fear of breaking it but I was surprised that it
didn't turn. Now that I think of it, I don't think that fan EVER
turned during the couple of hours I spent doing these tests. It didn't
dawn on me until now that it probably should have been turning all or
most of the time. (I told you I'm not a hardware guy! More proof of
why I'm not a hardware guy and shouldn't ever pretend to be ;-)

There is no wire running to the video card from elsewhere in the
computer. There is one pair of wires running from underneath the fan
to a spot a short distance from the fan on the card itself and that's
the only wire connected to the card. I tried powering up again (with
all four memory slots in use) and that video card fan is definitely
not turning. Before powering it up, I tried turning that fan with my
finger and pushed somewhat harder than I did yesterday; it's
definitely not turning. I wonder if it has a brake of some kind that
only releases when it has power?

Does the CPU fan turning off have significance in diagnosing the
problem?

Is there any more I can do for myself before bringing it in for that
diagnostic at the computer store? Are there any components we can be
sure have not failed, like the motherboard or CPU, or is the
diagnostic the best way of determining all that?

I'm trying to formulate alternatives. If I'm looking at building a
whole new computer (or pretty close), I wonder if I might be better to
remove the drives, mount them in an enclosure of some kind and access
them from my laptop and do without a desktop. That's got a major
problem associated with it - the desktop is XP SP3 and the laptop is
Windows 8 - but maybe I'd be able to see and access most of the data
anyway.

But I'm probably getting ahead of myself. Let's figure out how bad
this is first before declaring the desktop dead and moving on....

Paul

unread,

Apr 21, 2013, 6:55:55 PM4/21/13

to

I gave you my diagnosis, and I'll repeat it again.

1) B channel on CPU memory controller has failed. To regain
usage of all four slots for memory DIMMs, you'll need to test
with a replacement processor. If a test processor determines a working
A1/A2 channel and B1/B2 channel, then we know the original CPU
had a dead B channel. If for some reason the replacement processor
did not repair the problem, then it's motherboard. But the motherboard
contributes next to nothing, to memory operation on the M3A. Only
if a pin was bent, or the VDIMM regulator blew up, would the
replacement processor fail to fix it. You've already told me
a visual check of the DIMM slots reveals nothing. And your A1/A2
work, which proves VDIMM is working. Ipsofacto - replace CPU !
If you want to run with only two DIMM slots, use only A1/A2, then
hold off on buying the replacement CPU for now. We have to work
on our video. CPU is not a priority at the moment. But it has
to be fixed eventually, some day. We want working video, to prove
we're making progress.

2) The BIOS beep pattern is beeping about a video problem. You
have discovered a video card fan which no longer rotates. This
is a bad sign. Uh oh! In some cases, on fan failure, the heat from
the GPU, melts the plastic body of the fan (which is confirmation of
evil spirits :-) )

We term this a "double fault". And only testing, either temporary
testing (using computer shop test CPU and test video card), or you
purchase "worthy" replacement parts and test with them. If you do it
that way, you risk spending more for the overall repair, than going
through the computer shop. I repair my own systems... because
I'm an idiot, and I wouldn't do it any other way. My worst case
experience, I bought one new item, of everything. That's an expensive
way to do repair, but it's part of the gamble of repairing your own
stuff by yourself.

Most of the time, I narrow it down to just the defective item, but
I don't get lucky every time. Even a computer shop will guess incorrectly,
and sometimes it leads to the computer shop suffering burnt out test
components. That's the risk they take, and why the repair fee is higher
than it should be - it helps pay for the stuff they wreck.

*******

The bearing on the video fan has probably failed. The fan draws power
from the two pin header on the video card PCB. The video fan does not
need to be plugged into the motherboard, because there is a perfectly
good fan header right on the video card. It will be a two wire fan,
with no RPM measurement capabilities. Too bad the video card company
didn't spend a few cents on a fan failure indicator!

There is no point "just replacing the fan", because the card is
giving us evidence it is dead. We won't know it's really dead,
until a test card has been substituted in its place.

Now you need to borrow a video card for testing. With same slot
type as 9500 GT.

The idea would be, you test to see if you can enter the BIOS, or get
the BIOS screen up and running. If you boot into Windows with the new
card in place, you'll end up running at a low resolution (like 800x600),
depending on the driver situation. Some driver work may be needed,
to get full resolution back. Uninstall original driver from Add/Remove,
install new driver from video card boxed CD.

We don't know the video card is dead. If the card is naturally a low
power card, perhaps a fan failure won't kill it. On high power cards,
there have been cases where a fan failure, leads to the fan being
melted, as the chip underneath overheats :-)

If we look at your motherboard chipset info...

http://en.wikipedia.org/wiki/AMD_700_chipset_series#770

"770

One physical PCIe 2.0 x16 slot"

So you should be able to use any PCI Express video card you
can find in that slot. Subject to your power supply being
able to provide enough power. If you buy a $500 card, the
supply may not be adequate. If you buy (or borrow) a cheap
$50 card, the power is likely to be less than 50W.

For example, a Geforce 210 card is available for $30. Not
a gamer card by any stretch of the imagination. And your
local shop will likely charge a bit more than $30
for their Geforce 210. This is good enough to be a
"frame buffer", but may not be strong enough to run "The SiMMs"
latest version.

http://www.newegg.com/Product/Product.aspx?Item=N82E16814134157

Good luck,
Paul

hugh....@hotmail.com

unread,

Apr 21, 2013, 8:54:52 PM4/21/13

to

On 21 Apr, 18:55, Paul <nos...@needed.com> wrote:

You did. I suppose I'm saying I didn't really understand it. ;-)

> 1) B channel on CPU memory controller has failed. To regain
> usage of all four slots for memory DIMMs, you'll need to test
> with a replacement processor. If a test processor determines a working
> A1/A2 channel and B1/B2 channel, then we know the original CPU
> had a dead B channel. If for some reason the replacement processor
> did not repair the problem, then it's motherboard. But the motherboard
> contributes next to nothing, to memory operation on the M3A. Only
> if a pin was bent, or the VDIMM regulator blew up, would the
> replacement processor fail to fix it. You've already told me
> a visual check of the DIMM slots reveals nothing.

I don't remember doing any formal inspection. I certainly didn't look
at them carefully. I _did_ find that the memory cards went into some
slots more easily than others. Some were smooth as silk but others
needed some extra firmness and felt like there might be a bit of grit
in the track somewhere.

I just did a thorough inspection of all four DIMM slots and the memory
cards in them and can't see anything amiss in any of them. There was
one speck of dust in the B2 slot but a little puff of air blew that
away. The A2 slot was the one that took a little extra work to get the
card settled in.

> And your A1/A2
> work, which proves VDIMM is working. Ipsofacto - replace CPU !
> If you want to run with only two DIMM slots, use only A1/A2, then
> hold off on buying the replacement CPU for now. We have to work
> on our video. CPU is not a priority at the moment. But it has
> to be fixed eventually, some day. We want working video, to prove
> we're making progress.
>

So you think we might have a working computer with only the original
memory cards if we can get some video working to prove it. That sounds
encouraging!

> 2) The BIOS beep pattern is beeping about a video problem. You
> have discovered a video card fan which no longer rotates. This
> is a bad sign. Uh oh! In some cases, on fan failure, the heat from
> the GPU, melts the plastic body of the fan (which is confirmation of
> evil spirits :-) )
>

And you think this fan failed because of my mishap with changing the
memory with the power still on? Why would that affect the video fan?

> We term this a "double fault".

Does that mean we had two unrelated failures at the same time?

Coincidences do happen - I've seen quite a few - although I'm inclined
to think that two simultaneous failures with different causes seems
unlikely. But my knowledge of electronics is negligible so maybe this
is less implausible than it sounds to my ignorant ear.

In any case, I'm not all that concerned about why it happened. Maybe
the fan was already dead and has worked for weeks; I only noticed that
it wasn't turning as a result of trying to fix the other issue.

> And only testing, either temporary
> testing (using computer shop test CPU and test video card), or you
> purchase "worthy" replacement parts and test with them. If you do it
> that way, you risk spending more for the overall repair, than going
> through the computer shop. I repair my own systems... because
> I'm an idiot, and I wouldn't do it any other way. My worst case
> experience, I bought one new item, of everything. That's an expensive
> way to do repair, but it's part of the gamble of repairing your own
> stuff by yourself.
>
> Most of the time, I narrow it down to just the defective item, but
> I don't get lucky every time. Even a computer shop will guess incorrectly,
> and sometimes it leads to the computer shop suffering burnt out test
> components. That's the risk they take, and why the repair fee is higher
> than it should be - it helps pay for the stuff they wreck.
>

Agreed. They're not superhuman. They're bound to make mistakes too. I
was pleasantly surprised that they didn't laugh at me when I told them
what I'd done. Perhaps it was just them not wanting to piss off a
customer but I'd like to think it was because they've made a mistake
too on occasion.

> *******
>
> The bearing on the video fan has probably failed. The fan draws power
> from the two pin header on the video card PCB. The video fan does not
> need to be plugged into the motherboard, because there is a perfectly
> good fan header right on the video card. It will be a two wire fan,
> with no RPM measurement capabilities. Too bad the video card company
> didn't spend a few cents on a fan failure indicator!
>

It would only do you some good if the indicator appeared where you'd
see it, like an idiot light on your car's dash ;-)

> There is no point "just replacing the fan", because the card is
> giving us evidence it is dead. We won't know it's really dead,
> until a test card has been substituted in its place.
>
> Now you need to borrow a video card for testing. With same slot
> type as 9500 GT.
>

I've got previous computers but I don't imagine any of them would be
PCI Express. I've certainly never seen anything that looked like that
clamp arrangement on any of my previous PCs.

The guy at the computer store said he probably had a spare CPU that
would fit that mobo. I don't think he said anything about video cards
but there's at least a chance.

Maybe if I go there tomorrow and just see if he can swap out the video
card for me, he could do that while I wait - and free - rather than
putting it in his service queue....

I'm not much of a gamer. My main game is Hoyle Puzzle Games 2003,
which has Mahjong and other similar board and card games. ;-) I _do_
have Flight Simulator 2003 but haven't even tried it in years; I can
never get past a point in the third lesson before washing out.

But I do think I might upgrade to Windows 7 or 8 on the desktop so I'd
like a video card that is compatible with that.

For now, I'll see if I can get the guy at ComputersCanada to put in
another video card and see if it fires up with memory in the A1/A2
slots.

By the way, I said I assumed I was out of luck if I'd fried the new
memory and he said not necessarily since I'd only just purchased it.
He said if I went back to the manufacturer and said the right words,
they'd probably give me my money back. So maybe I'll give that a
whirl. If the B1/B2 slots aren't getting recognized anyway, I don't
need the extra memory unless I go for a new CPU.

Thanks very much for your help so far, Paul!! You've been great with
answering my dumbass questions and helping me go in a sensible
direction. I really appreciate that.

Paul

unread,

Apr 21, 2013, 10:26:39 PM4/21/13

to

hugh....@hotmail.com wrote:
> On 21 Apr, 18:55, Paul <nos...@needed.com> wrote:

>
>> We term this a "double fault".
>
> Does that mean we had two unrelated failures at the same time?

Yes.

Remember, we're relying on a sick computer, to tell us what's wrong.
Only you can listen to the beep pattern right now, and decide whether
that's a video beep error or not. Maybe the beep pattern is just
a fake, and the computer is crazy... :-)

That fan did not stop on its own. On my NVidia 7900GT, the fan
runs at 100% speed, until a Windows driver loads and slows it down.
(On Linux, it just wails at 100% the whole time. Only the
Nvidia tainted Linux driver will slow it down, under Linux.
As a hardware guy, I "fixed" this (evil grin). )

You don't have an OS running right now, which means the fan
should be at full power (such as it is). I'm not aware
of any NVidia video card implementations, with fully automated
fan control - such that a cool GPU would not require cooling
and the fan could actually stop rotating. Some motherboard/CPU
combinations work that way. There have been a few AMD processor
setups, where at startup, the CPU fan won't rotate. It has to
heat up a bit. Then it starts to spin.

>
> Coincidences do happen - I've seen quite a few - although I'm inclined
> to think that two simultaneous failures with different causes seems
> unlikely. But my knowledge of electronics is negligible so maybe this
> is less implausible than it sounds to my ignorant ear.

At the moment, I have no theory, to correlate your RAM installation,
finishing off the video card. I'm just dealing with the observations,
as thin as they are. I think you have a video beep error, you're
telling me the fan stopped spinning. That's not encouraging as
a set of symptoms. I'm trying to get us past the point, of being
stopped by a video beep error. If a substitute video card doesn't
do that, then we're back to changing out the CPU, and working on it
from that angle. (The theory being in that case, that the memory
operation is flaky enough, it's causing the video test to fail.)

Maybe you could blame a power supply failure, for causing the
symptoms, but then, we are able to get beep codes, and if
severe damage was done, we'd be getting no beeps at all.
The presence of any beeps, is a very encouraging sign,
as a lot of hardware on the core of the system has to
work, to get beeps out of it. It proves a lot of onboard
regulators have to be working.

<<snip>>

unread,

Apr 23, 2013, 1:53:22 PM4/23/13

to

On 22 Apr, 21:22, Paul <nos...@needed.com> wrote:

> hugh.mun...@hotmail.com wrote:
> > Why would my problem get worse? Is this
> > now a triple fault??
>
> > I have a strong suspicion that the next step is the fully $40
> > diagnostic and that we may now have THREE active problems:
> > a) the dead fan on the video card
> > b) the dead B channel for the memory, possibly attributable to the CPU
> > c) a dead power supply
>
> > This is starting to sound very expensive.....
>
> The $40 diagnosis, is beginning to sound pretty good about now.
> Head off to the store, and see what they say. If the repair
> list is too expensive, then bail.
>
> *******
>

I'm just back from taking the tower to the computer store. I went over
the situation with the same guy as before, whose name is Richard.
Unfortunately, he, had no recollection of our previous conversation
but gradually started to recall it as we went along.

Richard was really quite helpful and immediately started to try some
things to isolate the problem. I think he really wanted to figure out
what had failed so that he could sell me whatever I needed and get
back up and running immediately.

> All the fun, is in the guessing right now. Your diagnostics
> guy, will be able to figure this out for sure.
>
> I'm guessing +5VSB is being shorted by the motherboard, for
> some reason.
>

One of the first things he did was disconnect a large connector from
the mobo and attach it to a digital tester. (Don't ask me what the
tester is called; it looked like an oversized pager and gave digital
readouts on the display but that's all I can tell you.)

He tested 12 V first and said that part was fine, although the reading
may have been just a wee bit on the low side. I mentioned that you had
concerns about the 5 V circuit and he tested that next. He found the
numbers unreasonably low on the 5 V rail and seemed to be homing in on
the power supply as at least one of our culprits.

> I would pull the supply, put it on the bench, turn it on,
> then connect PS_ON# to COM with a jumper wire. (I have a
> home-made load tester, with a switch to do that.)
>

I thought about doing this but: a) I've never tried removing a power
supply and with all the wires coming out of it, found the prospect a
bit daunting for fear that I would not be able to put it back together
again
b) the only test equipment I have is one very cheap voltmeter
c) I'm not sure where to find PS_ON# or COM
d) I'm not sure what a jumper wire is or if I could construct one out
of materials at hand. I'm guessing it's just a normal length of wire
with insulation on it and bare ends, which I could find but if it's
something else, I might not have it

> My guess is, with no load, the +5VSB rail (the thing that
> runs the green LED), will come back. But, as soon as the
> power supply is connected to the motherboard, and the
> motherboard draws more than two or three amps from +5VSB,
> both the LED will be off and the supply (main section) will
> be off.
>
> I have a clamp-on DC ammeter, that I can test for that here.
> (Check the current draw on each rail.)
>

I think that's essentially what Richard did with his tester, as
reported above.

> The next test case, is placing a load on the supply that
> at least manages to spin its cooling fan. I have a power
> supply here, that passes the simple switch test (join
> PS_ON# to COM). But, as soon as you draw even a tiny bit
> of current from +12V (more than 0.1 amps), the +12V rail
> drops to nothing. If I were to connect my load box (maybe
> 100 watts of loading), that supply is useless. So there
> will be cases, where the problem shows ("weak supply"),
> only after you connect a representative load. The voltages
> from that supply are normal, when nothing is connected to
> it as a load.
>

I mentioned that the fan on the video card wasn't moving and couldn't
be made to turn by gentle finger pressure. He agreed that it was
seized. In answer to my question, he said that if I wasn't working the
card very hard, that fan could have seized months ago and not shown
any symptoms. He tried another video card and we were able to get into
the BIOS. He said the 5 V and 12 V numbers looked good in there.

However, we had a message about the CPU fan. He changed a bunch of
settings with regards to the CPU fan and then we got a lot of beeps on
the next bootup. He seemed confident that the CPU fan should be
replaced and showed me one that he said would be perfectly good for
$15. It was starting to look like a new video card and a new CPU fan
might be the extent of the problems. However, there were still doubts
about the power supply.

I gave him a printout of my diagnostic tests with the memory. We
discussed this and he was concerned about the B1/B2 DIMM slots. He
said there might be an issue with the motherboard that tied the
various problems together but that was not trivial to diagnose. When
he saw that the CPU fan WOULD turn just fine in cases where we got no
beep codes, I think he realized that things were more complicated than
he first hoped. I was certainly confused since he'd indicated possible
problems with the power supply, CPU fan, motherboard, as well as the
video card. So I suggested maybe the full diagnostic for $40 was the
way to confirm exactly which components had failed and he agreed.

So I left him the tower and memory and paid for the diagnostic. He's
going to try to have the results for me for Thursday but he's got
quite a lot in his queue so it might take longer.

He's even going to take a whirl at fixing the video card. I think he
wants to try cleaning and lubricating the fan to see if that gets it
behaving again.

> Your diagnostics guy will know how to do these tests,
> and come to some conclusion. Considering the number of
> things wrong with the system now, just the runtime alone
> for all the tests, is going to soak up the entire $40.
>

I suspect it's going to feel like money well-spent once we have a
relatively definitive set of results. Maybe, just maybe, I'll still
get out of this without having to spend a fortune....

I'll let you know what he reports. I'm sure you're curious to know
what the problem ultimately was ;-)
--
Rhino

Paul

unread,

Apr 23, 2013, 4:05:21 PM4/23/13

to

I think he's doing well so far.

About the only area he might be going off the rails, is with
regard to the B1/B2 symptoms.

An AMD motherboard, where the memory controller is inside the
processor, the motherboard contributes precious little to the
test results. The motherboard is just a bunch of copper wires
in that path. The motherboard copper wires, join the memory
controller inside the CPU, to the DIMM slots.

The fact that you got into the BIOS, means A1/A2 are working,
means A1/A2 had working VDIMM and Vtt supplies. In other words,
the motherboard has proved its merits.

His next step in resolving the B1/B2 problem, is to install
his test AM2 processor, and note that B1/B2 start working
again. Then, his conclusion will be, new CPU.

That's my best guess at this point.

*******

A computer motherboard is tremendously complex. Many kinds of
failures are theoretically possible (and I've seen a bunch
under lab conditions). The thing about production components,
motherboards made in the millions, a number of fault types
just don't happen. This is a function of the degree of control
the factory has, over the quality.

That leaves only a certain set of classes of failure, to be
seen in the field. 99% of the faults will fall into those
classes. For example, things exposed to the user, where
the user can "zap" things with static electricity, that
opens some fault possibilities. But those faults are "outward
facing", things like blown USB ports on a USB2 NEC chip.
Internal signals, like signals from Southbridge to Ethernet
controller, you never find those blown out. Similarly, you
never find cracked PCB copper traces. It just doesn't happen.
99.5% of the time, it's going to be something else which is
damaged, not the traces.

I had a poor quality PCB shop in town, which used to make
board blanks for us. We ordered small quantities from them.
A board blank came in, we built it up, soldered components
to it, tested it, stored it in a shelf PCB holder for later.
That board tested as good. Three months later, I tested it
again and it failed. My boss sent it out for bed-of-nails
testing, and the tester showed that over 200 copper traces
had gone open circuit. That means the blank PCB was defective,
(open circuits happened while the PCB was cold and stored
unpowered in the lab) and was fabricated improperly.
Now, our production boards, *never* do that. Neither do the
three million boards a month that Asus might make. The shop in
town was disqualified as a supplier, and we shopped elsewhere.

So when I listen to a set of symptoms here, I tend to dismiss
some fault types as very unlikely. In the case of your DIMM
problem, I see evidence the DIMMs are being powered OK. Since
the A1/A2 are working, that leaves connectivity on B1/B2
(bent pins, broken trace, or bad CPU). You've verified no bent
pins, the symptoms affect both B1/B2, and that says bad CPU.
As to when it blew, we'll never know. If could be, that the
B-bus was bad on the CPU from day one. But more likely,
something happened when you were plugging in DIMMs. So while
lots of things could happen (as in my example, broken traces),
the vast amount of evidence out there, is broken traces
just don't happen on production boards. Production systems
use closed loop feedback, to tune production and watch for
problems. For example, in our production, we got a bad
batch of boards, and it was detected by impedance coupon
testing, and the boards didn't even get loaded into the
pick and place machine. The boards were put aside, to be
cross-sectioned and checked. they were out of spec, and
the supplier ate the loss. They're actually supposed to
check the impedance coupon themselves, as part of their
checks.

*******

On the video card, I'd inspect the plastic fan body for
signs of melting. If it's melted, that means the chip
temperature went too hot. (The fact the bearing is seized,
means something, but I don't know what.) I don't see any
evidence right now, that the 9500 GT can be recovered. The
fact his shop video card worked, rules out a bad PCI Express x16
slot. (Note that, it's possible to blow lanes on an x16
slot, and the slot will still work. The negotiation just
narrows down the lane set, and you get reduced slot
bandwidth. What the slot can't handle well, is if lane 0
fails, then that would be deadly. Lane 15 isn't nearly as
important, and if lane 15 fails, it can run in x8 mode.)

Paul

hugh....@hotmail.com

unread,

Apr 23, 2013, 8:01:43 PM4/23/13

to

On 23 Apr, 16:05, Paul <nos...@needed.com> wrote:

I don't think he considers damaged traces very likely, more of an
outside possibility. He said he wasn't going to go over the whole mobo
with a voltmeter testing each circuit. For one thing, he said he
doesn't have a voltmeter ;-) But mostly the issue is time; checking
every circuit would take a LONG time. Basically, he said it just isn't
done; instead, the mobo is replaced if it turns out to be the
problem.

> I had a poor quality PCB shop in town, which used to make
> board blanks for us. We ordered small quantities from them.
> A board blank came in, we built it up, soldered components
> to it, tested it, stored it in a shelf PCB holder for later.
> That board tested as good. Three months later, I tested it
> again and it failed. My boss sent it out for bed-of-nails
> testing, and the tester showed that over 200 copper traces
> had gone open circuit. That means the blank PCB was defective,
> (open circuits happened while the PCB was cold and stored
> unpowered in the lab) and was fabricated improperly.
> Now, our production boards, *never* do that. Neither do the
> three million boards a month that Asus might make. The shop in
> town was disqualified as a supplier, and we shopped elsewhere.
>

PCB is the material the circuit board is made of, right? The basic
green plastic? Poly-something-something? My knowledge of electronics
and the techniques used to make chips and boards is negligible. I
projected a film that talked about the process once when I had a part-
time gig showing films on campus and that's been pretty much my whole
exposure to electronics, I'm afraid ;-) And that was 30+ years ago so
even that isn't very up-to-date ;-)

> So when I listen to a set of symptoms here, I tend to dismiss
> some fault types as very unlikely. In the case of your DIMM
> problem, I see evidence the DIMMs are being powered OK. Since
> the A1/A2 are working, that leaves connectivity on B1/B2
> (bent pins, broken trace, or bad CPU). You've verified no bent
> pins, the symptoms affect both B1/B2, and that says bad CPU.
> As to when it blew, we'll never know. If could be, that the
> B-bus was bad on the CPU from day one. But more likely,
> something happened when you were plugging in DIMMs. So while
> lots of things could happen (as in my example, broken traces),
> the vast amount of evidence out there, is broken traces
> just don't happen on production boards. Production systems
> use closed loop feedback, to tune production and watch for
> problems. For example, in our production, we got a bad
> batch of boards, and it was detected by impedance coupon
> testing, and the boards didn't even get loaded into the
> pick and place machine. The boards were put aside, to be
> cross-sectioned and checked. they were out of spec, and
> the supplier ate the loss. They're actually supposed to
> check the impedance coupon themselves, as part of their
> checks.
>

No, that makes perfect sense. Look for things that actually happen in
the real world; don't worry about things that are only theoretically
possible but never seen.

> *******
>
> On the video card, I'd inspect the plastic fan body for
> signs of melting. If it's melted, that means the chip
> temperature went too hot. (The fact the bearing is seized,
> means something, but I don't know what.) I don't see any
> evidence right now, that the 9500 GT can be recovered. The
> fact his shop video card worked, rules out a bad PCI Express x16
> slot. (Note that, it's possible to blow lanes on an x16
> slot, and the slot will still work. The negotiation just
> narrows down the lane set, and you get reduced slot
> bandwidth. What the slot can't handle well, is if lane 0
> fails, then that would be deadly. Lane 15 isn't nearly as
> important, and if lane 15 fails, it can run in x8 mode.)
>

I expect Richard will do something along those lines. I was a little
suprised that he even considered fixing the card. He'd earlier said
you can't really fix them and pointed out that a decent replacement
card (I don't recall which one) was only about $45. Then suddenly he
talked about repairs. I suspect he was really hoping to diagnose the
whole problem with what he did today and is offering to try fixing the
video card as sort of a consolation for him having to ding me $40 for
the diagnostic....

In any case, I have reasonable comfort that he is competent which is a
big hurdle. It wouldn't be hard for him to tell me a bunch of BS but I
think he was actually being straight with me.

Here's hoping he is!

--
Rhino

hugh....@hotmail.com

unread,

Apr 26, 2013, 8:08:20 PM4/26/13

to

Just to update this, I just got off the phone with the tech at the
store. He says he's pretty sure it is the motherboard that is the main
problem. The video card is also gone but we knew that already. The 2nd
and 4th DIMMs don't seem to work with either my memory or other memory
he has that he knows is good. The power supply is a little bit weak on
the 3.3 line and he gets CPU fan failures even with a brand new power
supply.

He's proposing a replacement motherboard that is also AM2 and holds
either 2 DDR2 chips or 2 DDR3 chips (but not both at the same time).
He says some of the stores stock it. He gave me their part number and
told me how to determine which stores have it in case I wanted to grab
it myself. Then he ended the call because it was the end of his day.

And here's where Murphy's Laws come into play. I looked up this item
number in the way he suggested and it says there are none of these
boards anywhere in their chain. Zero, zilch, nada. I tried to call
back but it was already past closing time and no one picked up. I know
he's not there this weekend. That means I can't tell him that there
are no copies of the mobo in their system (or find out that I am doing
the search incorrectly) until Monday and that means there will be a
further delay while we figure out another alternative and order it
in.

I hope I'm just doing something wrong because this new board sounded
like it would really solve my problems. It even has VGA on it which
might give me a picture that I could live with so that I didn't have
to buy another video card. The price is certainly right, about $45
plus about $10 for installation. He said he could also arrange a
return on the 2 GB of memory that I bought since it is less than 2
weeks since I bought it.

The alternative is to get a new motherboard, CPU, and memory but he
ballparked that at $250-300 depending on what I wanted.

I'm POSITIVE I wrote down the item number correctly. I read it back to
him phonetically and he said I had it right. MBAR000445. I suppose
that is just an internal number within the store so probably doesn't
help me locate it anywhere else. I'll look after I've fed the cats. If
I don't find it, I'll likely be back asking for suggestions.

hugh....@hotmail.com

unread,

Apr 26, 2013, 8:28:25 PM4/26/13

to

I think I know a part number that means something outside of Canada
Computers. I googled on MBAR000445 and got hits that mention both
MBAR000445 and ASRock N68C-GS FX. That board is described here:
http://www.asrock.com/mb/NVIDIA/N68C-GS%20FX/

Any thoughts on whether this would be a good replacement for the M3A
board?

Or would I really be better to get a new machine built? I would
strongly prefer to keep my costs down so this ASRock board is looking
awfully tempting, assuming I can actually get my hands on it. I'm
going to see if I can track it down anywhere in my area.

Paul

unread,

Apr 26, 2013, 9:33:14 PM4/26/13

to

I can see the same board here. Two DDR2 DIMM slots. Two DDR3 DIMM slots.
That means two DIMMs max at any one time, as you cannot mix the two
types at the same time. So you'll never be able to use four DIMMs
that way. Chipset is NVidia.

http://www.ncix.com/products/?sku=71223&vpn=N68C-GS%20FX&manufacture=ASRock

An alternative is this ASRock A785GM-LE, but it's only got
two DIMM slots as well. Chipset is ATI (AMD).

http://www.newegg.com/Product/Product.aspx?Item=N82E16813157274

Good boards are just hard to find. This is generally the way
it happens too - the remaining boards at retail, only
have two DIMM slots. If you're happy with that, then fine.

Before purchasing any motherboard, always check the CPU support
chart. Is your processor here ? If the VCore regulator is
under-powered, the CPU support chart will help you detect
that fact. If your CPU isn't there, then the VCore might be
too weak. That's just one of the reasons for doing a check
here. The needed BIOS revision is the other reason. You'll
notice the majority of entries in the table, have a BIOS
revision of "All", meaning support was there from day one.

http://www.asrock.com/MB/overview.asp?cat=CPU&Model=N68C-GS%20FX

*******

To reach the conclusion he has, he should have:

Inserted your CPU, into a known good AM2 motherboard,
and verified A/B memory channels work. If dual channel
mode works there, he can say "it's your motherboard".

If he inserted a known good CPU into your motherboard,
and B channel started working, he could say "it's your CPU".

I would do the known good AM2 motherboard test first, assuming
he has one.

Without that test, he's just guessing. Just like me. You need
sufficient known-good test gear, to be in the testing business.

At the very least, get your money back on the RAM. Then
consider your options. For example, have a look around
Ebay. This one is from China, but perhaps you can find
something a little closer to home.

http://www.ebay.com/itm/ASUS-M3A78-AMD770-SB700-AM2-AM2-Phenom-DDR2-motherboard-Free-shipping-/300888467044?pt=Motherboards&hash=item460e59a664

If you go to support.asus.com, you can do a "search by CPU",
to find a list of motherboards that will work.

http://support.asus.com/CPUSupport/List2.aspx?SLanguage=en

Select your CPU from the drop down menus, and see if the
M3A78 is a candidate or not. That will give you a
good-sized list to start with, while flipping through
Ebay listings.

Paul

hugh....@hotmail.com

unread,

Apr 27, 2013, 7:58:23 AM4/27/13

to

On 26 Apr, 21:33, Paul <nos...@needed.com> wrote:

> http://www.ncix.com/products/?sku=71223&vpn=N68C-GS%20FX&manufacture=...

> http://www.ebay.com/itm/ASUS-M3A78-AMD770-SB700-AM2-AM2-Phenom-DDR2-m...

>
> If you go to support.asus.com, you can do a "search by CPU",
> to find a list of motherboards that will work.
>
> http://support.asus.com/CPUSupport/List2.aspx?SLanguage=en
>
> Select your CPU from the drop down menus, and see if the
> M3A78 is a candidate or not. That will give you a
> good-sized list to start with, while flipping through
> Ebay listings.
>
> Paul

Thanks for the suggestions, Paul. I think I'm going to be following
them very closely.

It's just my luck that the tech is off today and tomorrow. It's going
to make it difficult to determine if he did the test of the CPU. He
didn't specifically mention that last night. I know he INTENDED to do
that test but I don't remember him saying specifically that he did. I
think he may have been doing some kind of written test result -
although he didn't say that specifically either - so I'll call when
they open and see if his colleagues can find that; it should say if he
tested my mobo with a known-good CPU. I'll also ask them if they can
track down the recommended MOBO either in their system or by ordering
it from ASRock.

In the meantime, I can start following your suggestions. Maybe I can
find something suitable via my own searches....

Paul

unread,

Apr 27, 2013, 9:59:52 AM4/27/13

to

If you are happy with having two DIMM slots, then one of the boards
at retail would be fine. It's if you want four DIMM slots, like
the original motherboard, have a look around Ebay. And preferably,
deal with someone in a country, where you have some legal leverage.

The reason for asking a few more questions of your tester,
is to make sure that shopping for the motherboard is absolutely necessary.
If he admits there is still some uncertainty, then at least
you'll know, when the new motherboard comes in and you do your first test.

The motherboard contribution to the defective B1/B2 consists of:

1) CPU socket pins. An open or short there could break B1/B2.
As far as I'm concerned, ZIF sockets seem to have excellent
reliability. Even when abused (pull CPU out of socket),
they still work. They're slightly better than LGA, in that
LGA sockets, the pins sometimes get crushed in transit.

2) Copper tracks on motherboard cracked or open circuit.
Again, generally speaking, the motherboards don't have a problem
like that. With the new solder they use, sometimes motherboards
have solder-ability problems. And you can certainly peel a track
off a motherboard with your soldering iron - motherboards aren't
"highest quality" when it comes to heat resistance. But assuming
your B1/B2 worked at some time, I don't see a reason to assume
a fault like this.

If you're using a tool and the tool slips, and rips or scratches
something on the board, that can do it. I've had people on the
newsgroup, who have managed to bust SMT resistors off the motherboard.
It's real easy to do that, if you're using pliers to modify a
push pin. I've had a couple near-misses.

3) Broken connection at B1 and/or B2 DIMM slots.

4) On DDR motherboards, there were resistors for SSTL termination.
An example here, of the termination scheme.

http://www.maximintegrated.com/app-notes/index.mvp/id/1775

On DDR2, they use on-die termination. So you can't bust those off.
Thus DDR2 has slightly better reliability. (I've been looking for
some pictures like this...)

http://web.archive.org/web/20060102003658/http://www.elpida.com/pdfs/E0593E10.pdf

The VDIMM and VTT regulators are proven, by the A1/A2 tests passing.
All the slots share those facilities. There's very little else
that can go wrong with it.

If he tested your CPU in a known-good motherboard, and B1/B2
were working, then it's time for another motherboard.

Paul

hugh....@hotmail.com

unread,

Apr 27, 2013, 10:21:37 AM4/27/13

to

On 26 Apr, 21:33, Paul <nos...@needed.com> wrote:

> http://www.ncix.com/products/?sku=71223&vpn=N68C-GS%20FX&manufacture=...

> http://www.ebay.com/itm/ASUS-M3A78-AMD770-SB700-AM2-AM2-Phenom-DDR2-m...

>
> If you go to support.asus.com, you can do a "search by CPU",
> to find a list of motherboards that will work.
>
> http://support.asus.com/CPUSupport/List2.aspx?SLanguage=en
>
> Select your CPU from the drop down menus, and see if the
> M3A78 is a candidate or not. That will give you a
> good-sized list to start with, while flipping through
> Ebay listings.
>
> Paul

Good news. I called the store the minute it opened and talked to the
tech on duty. It turns out that he'd actually been briefed by Richard
while he was working on my computer. Also, Richard kept some written
notes for the report on my problem. Now, the guy I spoke to this
morning sees nothing in the notes about Richard specifically testing
with a different CPU but is pretty sure that he did. He says Richard
won't normally order new parts unless he's sure what the problem is.
Also, he says CPU failures are extremely rare; he's only seen 3 in his
13 years as a tech. But since he doesn't see it in the notes, he's
either going to call Richard at home and verify that he did that test
or he is going to test the CPU himself today to verify that my CPU is
still good.

More good news. Although I can't find the specified motherboard in
their system, it has in fact been ordered and the tech is sure that
the board is on its way. He says that when they have an item that has
very low stock numbers, it will often be removed from the website but
their internal system still shows that another store has it. There's a
remote chance the board will be in by Monday but it's more likely I'll
get it Wednesday. (They only get deliveries Mondays, Wednesdays, and
Fridays.)

So it's a lot less likely now that this is going to be a case of
waiting for several days for a replacement part that won't solve the
problem.

I also feel a little better because the guy I spoke today had done the
same thing I had done once and hot-swapped some memory. If a
professional tech can do that - mind you, I don't know if he did this
before he'd been trained - then I don't feel like such an idiot.

--
Rhino

Paul

unread,

Apr 27, 2013, 10:43:54 AM4/27/13

to

hugh....@hotmail.com wrote:

>
> I also feel a little better because the guy I spoke today had done the
> same thing I had done once and hot-swapped some memory. If a
> professional tech can do that - mind you, I don't know if he did this
> before he'd been trained - then I don't feel like such an idiot.
>
> --
> Rhino

Well, lots of times I forget to remind people about turning
off the power. So it's an easy detail to miss.

My Asrock board doesn't have the green LED on it, so
I have to be extra careful with that one, when working
inside the box. No green LED, means I have no warning
to work with. Asus is the brand that likes the green LED.

Of course, you can always solder your own green LED and
limiting resistor to the motherboard. So a warning
LED can easily be retrofitted, if you want one.
I keep a selection of LEDs here, for random
indicator usage. In the last year, I bought around
150 LEDs. So there's always a few kicking around.

Paul

hugh....@hotmail.com

unread,

unread,

May 12, 2013, 2:30:40 PM5/12/13

to

http://www.asrock.com/mb/NVIDIA/N68C-GS%20FX/

"5.1 CH HD Audio (VIA VT1705 Audio Codec)"

ftp://174.142.97.10/manual/N68C-GS%20FX.pdf

And it would be connected to Nforce 630a.

I can see a VIA audio driver here, for the VT1705.
The driver is for multiple chips. It's approximately 96MB.

http://www.asrock.com/mb/NVIDIA/N68C-GS%20FX/?cat=Download&os=XP

ftp://174.142.97.10/Drivers/All/Audio/VIA_Win7-64_Win7_Vista64_Vista_XP64_XP(v9700a).zip

When you open that ZIP file with 7ZIP, you can find
these auxiliary files for handling any missing Microsoft UAA.
This is just to explain how audio drivers handle UAA, when
the user didn't already fix it themselves.

VIAHDAud\
HDBusDrv\
HDWXPx64.exe (WinXP 64 bit edition)
HDWXPSp2.exe <--- WinXP patched to SP2
HDWXPSp1.exe <--- WinXP patched to SP1
HDW2K3x64.exe
HDW2K.exe
HDSrv2K3Sp1.exe
HDSrc2K3.exe

There is no UAA driver for WinXP patched to SP3, because
SP3 includes its own copy. If you installed SP3, you could get
the UAA that way.

If you look in Device Manager (Start : Run : devmgmt.msc),
you should see this in the System section.

Microsoft UAA Bus Driver for High Definition Audio

Now, I don't even know how I got mine :-( My OS disc has
been used on two motherboards, and I can't tell where the UAA
driver came from. Since it is a "native" SP3 system, it
makes sense that UAA was already installed. And so, my
UAA entry was there from day one. And that means I have no
Add/Remove for it. If I did need to reinstall it for some
reason, I could do "Driver Update", navigate through the
drivers on my WinXP CD (or equivalent), as there is an
HDAUDBUS.IN_ file on there. When that file is decompressed,
the name changes to HDAUDBUS.INF and that in turn uses
HDAUDBUS.SYS.

OK, so did you uninstall the previous driver ?

Take a look in Add/Remove Programs and uninstall the old driver.
(The old HDAudio was RealTek ALC883 on Asus M3A motherboard.)
If the old audio codec wasn't VIA, maybe it would not conflict at
all, so in this case, I don't see why that would be causing
a problem. Unless, like my 8738 driver from long ago, it
leaves services running (that launch, even when the underlying
hardware is not present!). I had a stupid sound card that did that.
That's a reason to remove the old driver, to dump the old service
associated with it. You don't want services or control panels
from old (non-existent) sound cards to be running - no end of
troubles.

When you use the setup.exe on the new driver, it will call up
HDWXPSp1.exe, HDWXPSp2.exe, or no UAA driver at all if the
OS is at SP3. And then, the installation for the VIA chip
can go from there. The VIA chip should be enabled in the BIOS.
In the motherboard manual you will see -

Section 3.4.2 of manual...

Chipset Settings:

Onboard HDAudio [Auto]
Front Panel [Auto]

If the HDAudio entry was disabled, that would cause a problem.

The Asus tech support recommendation, for UAA problems is
as follows. You only need to investigate this, if
removing the RealTek from Add/Remove, then installing
the VIA, still isn't working. You might do this,
if a "Code" was showing for the UAA device in System section
of Device Manager.

*******
1)Go in safe mode [Press F8 at startup, boot into Safe Mode]

2)Disable the HD audio device under Sound Video Games
controllers, do not uninstall it here! [May be no entry at this point]

3)Disable under the system devices "Microsoft UAA bus driver for high
definition audio", do not uninstall it here!

4)Now uninstall the "Microsoft UAA bus driver for high definition audio"
(via this device manager in safe mode), the HD audio device will then
also be uninstalled automatically

5)Install the latest driver for your motherboard. {That would be your
VIA driver]

In case you can not uninstall the UAA driver, go to microsoft.com and
find KB888111. It has some regedit instructions to remove KB888111.
[References to KB888111 might not make sense for a SP3 user, as
KB888111 was for the other service packs as far as I know. On SP3,
you could try installing a driver via Update Driver and use the driver
on the WinXP disc.]
*******

Very confusing procedure... but I've been flipping back and forth
to this for the last couple hours.

Go back to Add/Remove Programs, and see if the Disc Wizard software
created a separate entry for the driver. Try removing that virtual
disk driver. Then, go to Disc Wizard, and use the Allocate thing,
as I explained in my other post, and see if Disk Wizard will reinstall
the driver. After a reboot, all three partitions should show up.

That driver must have a "trick" as to how it connects to the existing
disk driver. And since you've changed Southbridges, and Southbridge
drivers, the virtual disk driver didn't "move" on its own. It needs
to be uninstalled and reinstalled. If the file was available separately,
that would be a bit easier.

The Disk Wizard is a re-branded Acronis product. The instructions
here, show the extended capacity manager (i.e. driver) can be removed.
And then you should be able to install it again. The interface is similar
to what the Disc Wizard flash video shows.

http://www.acronis.com/support/documentation/ATIH2012/index.html#15443.html

And apparently, the virtual disk driver is available separately,
in response to a request to install it manually. I downloaded this,
for a look at the driver files, to see what it is doing.

http://kb.acronis.com/content/38937

(1.8MB)
http://kb.acronis.com/system/files/content/2013/01/38937/virtualdisksetup.zip

Now, I look at things like that, to see how they work.

There are two interesting files in there.

Binary._2409C8CDE6643FE2ED727EE9BB3A52A8 is for 64 bit OSes.
Binary._847C629B04FEC51C40D8CCEDBD6BF452 is for 32 bit OSes.

When you use 7ZIP and "Open inside", these are the files. I'll
use the 32 bit version as an example.

Directory of C:\Downloads\Acronis_extended_capacity_manager_virtual_3TB

08/16/2012 06:31 PM 1,561 vididr.inf
08/16/2012 06:31 PM 399 vididr_uninstall.inf
08/16/2012 06:31 PM 8,130 vididr.x86.cat
08/16/2012 06:31 PM 139,336 vididr.x86.sys
08/16/2012 06:34 PM 1,457 vidsflt.inf
08/16/2012 06:34 PM 376 vidsflt_uninstall.inf
08/16/2012 06:34 PM 1,216 vidsflt_update.inf
08/16/2012 06:34 PM 7,515 vidsflt.x86.cat
08/16/2012 06:34 PM 99,720 vidsflt.x86.sys
08/16/2012 06:43 PM 18,558 vdd_msi_32.msg

12 File(s) 1,086,340 bytes

So those are the actual files used. If I open "vidsflt.inf", I
can see stuff like:

[FltServiceRegInfo]
HKLM, "System\CurrentControlSet\Control\Class\{4D36E967-E325-11CE-BFC1-08002BE10318}",
"LowerFilters", 0x00010008, %FltServiceName%

And I can find that GUID number, in this key.

HKEY_LOCAL_MACHINE\SYSTEM\ControlSet001\Control\Class\{4D36E967-E325-11CE-BFC1-08002BE10318}

and there are entries for perhaps every disk I've ever installed. The
main key mentions "vidsflt" and that's the thing that Acronis is
accessing/updating. So this is a "filter" driver, that intercepts
calls to the SATA port, and "makes" the virtual drive for you.
And it must be bound to the appropriate SATA port (which in your
case, is brand new, and the registry entry is missing).

It looks to me, like moving a disk around, required the driver
to be reinstalled, so that the "LowerFilter" gets put in the proper
place. It's just a matter of how to get that driver reinstalled.
The Extended Capacity Manager apparently has a button to remove it,
and then you could apply it again. That should only affect the
driver, and not write to the disk. What writes to the disk, is
things like a visit to Disk Management, and doing "partition type
stuff" and "formatting". That would mess it up.

Have I ever mentioned backups before... :-) Any time you budget
for a new disk, you should be buying two. The second disk,
holds your backup tool images of the contents of the disk.
If you have a backup, then you never have to worry. Now,
preparing for the mess you're in here, isn't easy. I don't know
if there is a way to set up a second 3TB drive, that would not
be using the services of the Acronis virtual driver. I would
be looking for some way to avoid using that. Maybe the services
of two 2TB drives, could provide backups for the three partitions
on the 3TB drive.

If you had backups, you would not have to worry about trashing
anything. I don't think there is anything to worry about in
this case, but what do I know :-(

Paul

Henry

unread,

May 14, 2013, 1:45:21 PM5/14/13

to

That's the file I downloaded.

>
> When you open that ZIP file with 7ZIP, you can find
> these auxiliary files for handling any missing Microsoft UAA.
> This is just to explain how audio drivers handle UAA, when
> the user didn't already fix it themselves.
>
> VIAHDAud\
> HDBusDrv\
> HDWXPx64.exe (WinXP 64 bit edition)
> HDWXPSp2.exe <--- WinXP patched to SP2
> HDWXPSp1.exe <--- WinXP patched to SP1
> HDW2K3x64.exe
> HDW2K.exe
> HDSrv2K3Sp1.exe
> HDSrc2K3.exe
>
> There is no UAA driver for WinXP patched to SP3, because
> SP3 includes its own copy. If you installed SP3, you could get
> the UAA that way.
>

I have exactly the files you listed.

My system was originally XP with no service packs. I installed SP2 some
years back and didn't go to SP3 immediately because I kept hearing
remarks about problems with it. But when I bought the 3TB hard drive in
December 2011, I found that I had to have SP3 in order to be able to
access all of the space on the drive so I installed SP3 then.

> If you look in Device Manager (Start : Run : devmgmt.msc),
> you should see this in the System section.
>
> Microsoft UAA Bus Driver for High Definition Audio
>

Yes, it's there. It looks fine too: no warning/error flags beside the
entry. The Properties dialog says it is enabled and working and has no
IRQ conflicts.

> Now, I don't even know how I got mine :-( My OS disc has
> been used on two motherboards, and I can't tell where the UAA
> driver came from. Since it is a "native" SP3 system, it
> makes sense that UAA was already installed. And so, my
> UAA entry was there from day one. And that means I have no
> Add/Remove for it. If I did need to reinstall it for some
> reason, I could do "Driver Update", navigate through the
> drivers on my WinXP CD (or equivalent), as there is an
> HDAUDBUS.IN_ file on there. When that file is decompressed,
> the name changes to HDAUDBUS.INF and that in turn uses
> HDAUDBUS.SYS.
>
> OK, so did you uninstall the previous driver ?
>

I didn't touch this UAA Bus Driver for High Definition Audio. Ever.

But when I tried to re-install the flagged audio driver in the "Sound,
video and game controllers" section, it wouldn't let me while the
existing driver was there so I uninstalled it via Device Manager and
then tried to install the VIA HD Audio Codec Driver via the Setup.exe
file in the zip file, it failed. The installation fails identically when
I use the CD that came with the mobo and choose "HD Audio Driver and
Application". Am I simply doing the wrong thing? Should I be selecting
something else like "All in 1 Driver" from the CD? Or some other file or
setup routine from the zip file?

> Take a look in Add/Remove Programs and uninstall the old driver.
> (The old HDAudio was RealTek ALC883 on Asus M3A motherboard.)

My entry doesn't say ALC883, it says "Realtek High Definition Audio
Driver" but I assume that is what you wanted me to remove so I
uninstalled it. I didn't reboot yet because I want to finish this email
first.

> If the old audio codec wasn't VIA, maybe it would not conflict at
> all, so in this case, I don't see why that would be causing
> a problem. Unless, like my 8738 driver from long ago, it
> leaves services running (that launch, even when the underlying
> hardware is not present!). I had a stupid sound card that did that.
> That's a reason to remove the old driver, to dump the old service
> associated with it. You don't want services or control panels
> from old (non-existent) sound cards to be running - no end of
> troubles.
>
> When you use the setup.exe on the new driver, it will call up
> HDWXPSp1.exe, HDWXPSp2.exe, or no UAA driver at all if the
> OS is at SP3. And then, the installation for the VIA chip
> can go from there. The VIA chip should be enabled in the BIOS.
> In the motherboard manual you will see -
>
> Section 3.4.2 of manual...
>
> Chipset Settings:
>
> Onboard HDAudio [Auto]
> Front Panel [Auto]
>
> If the HDAudio entry was disabled, that would cause a problem.
>

Okay, looks like I'd be best to reboot at this point. I'll save this
post as a draft and return to it once I've rebooted.

> The Asus tech support recommendation, for UAA problems is
> as follows. You only need to investigate this, if
> removing the RealTek from Add/Remove, then installing
> the VIA, still isn't working. You might do this,
> if a "Code" was showing for the UAA device in System section
> of Device Manager.
>

After the reboot, I tried re-installing the "HD Audio Driver and
Application" from the CD and it failed again, exactly as before. Then I
tried it from the SETUP.EXE in the downloaded zip file and it failed
again in exactly the same way. Device Manager continues to show a
warning flag on "Audio Device on High Definition Audio Bus" in Sound,
video and game controllers; Device Status still says "The drivers for
this device are not installed [Code 28]". Microsoft UAA Bus Driver for
High Definition Audio continues to look just fine.

I assume that means I should NOT do the steps listed next:

> *******
> 1)Go in safe mode [Press F8 at startup, boot into Safe Mode]
>
> 2)Disable the HD audio device under Sound Video Games
> controllers, do not uninstall it here! [May be no entry at this
> point]
>
> 3)Disable under the system devices "Microsoft UAA bus driver for high
> definition audio", do not uninstall it here!
>
> 4)Now uninstall the "Microsoft UAA bus driver for high definition audio"
> (via this device manager in safe mode), the HD audio device will then
> also be uninstalled automatically
>
> 5)Install the latest driver for your motherboard. {That would be your
> VIA driver]
>
> In case you can not uninstall the UAA driver, go to microsoft.com and
> find KB888111. It has some regedit instructions to remove KB888111.
> [References to KB888111 might not make sense for a SP3 user, as
> KB888111 was for the other service packs as far as I know. On SP3,
> you could try installing a driver via Update Driver and use the driver
> on the WinXP disc.]
> *******
>
> Very confusing procedure... but I've been flipping back and forth
> to this for the last couple hours.
>

I'm going to stop this post here to give you a chance to react.

I'll reply to the rest of the post in a separate reply.

For what it's worth, a new problem has emerged, although I can't think
why it would have anything to do with the mobo swapout. I was originally
replying to the post on my desktop but when I clicked on Send, nothing
happened. Ditto for the menu options Send Now and Send Later. I'm using
Thunderbird 17.0.6 as both my email client and my newsreader. Send works
fine in email mode but stubbornly refuses to work in newsgroup mode. I
think that was the first post since my mobo swapout but I'm darned if I
can figure out why that would cause this weird behaviour from
Thunderbird. I'm just mentioning it in case it offers a clue to the
other problems.
--
Rhino

Paul

unread,

May 14, 2013, 2:53:16 PM5/14/13

to

Either there is some part of the RealTek driver that is still present.
(Reboot, to clear out things like the old audio service.)

Or, the problem is related to UAA. But if you're at SP3, then
if you remove UAA, the OS should reinstall it as far as I know.

You can try the five step procedure in Safe Mode. I would expect
if you removed UAA, on the next reboot into regular mode, the UAA
will install itself again.

If for some reason, it did not, you could try reinstalling
it from the drivers Windows already has. (Try update driver
or scan for new hardware, that sort of thing.)

Paul

Rhino

unread,

May 17, 2013, 6:14:24 PM5/17/13

to

Sorry for the delay in following up on this thread. The pain from
banging my head against the wall was so intense that I had to step
away from the problem for a while. One bit of good news: the problem
with my Send button in composing newsgroup posts turns out to have
been unrelated to the motherboard issue.

Now, back to the audio problem.

I took another run at the issue today and tried the steps you suggested
for safe mode. I disabled the HD Audio Device in Sound, Video, Game
Controllers. I disabled Microsoft UAA Bus Driver. I uninstalled
Microsoft UAA Bus Driver. I tried to install the drivers from the CD but
got "InstallShield - 1607: Unable to install InstallShield Scripting
Runtime". I tried two or three times but got that error each time so it
clearly wasn't able to run in safe mode for some reason.

I went back to normal mode with the intent of reinstalling from there.
As I booted up, the Found New Hardware Wizard tried to install "MEDIA"
but I got this message: "Cannot install this hardware. There was a
problem installing MEDIA. An error occurred during installation of this

device. The name is already in use as either a service name or a service
display name."

I went into Control Panel/Sounds and Audio and found that there is still
"no audio device" in the Volume tab and no default device in Audio or
Voice tabs. The Microsoft UAA Bus Driver entry in Device Manager/System
Devices looks perfectly fine. There is an "Unknown Device" in Sound,
Video and Game Controllers. Its properties show Device Status as
follows: The drivers for this device are not installed. (Code 28). To
reinstall the drivers for this device, click Reinstall Driver." As if I
haven't already tried that a dozen times or more.

What the heck is going on? Is the audio circuit defective in this board?
Or have we got some kind of software/firmware issue here?

Am I going to have to go out and buy a sound card for this thing to get
my audio to work again? If so, what type of card do I buy? The mobo
manual refers to: PCI Slots (PCI1-2); PCI Express x16 Slot (PCIE2); PCI
Express x1 Slot (PCIE1) so I'm not sure just what I should look for if a
separate sound card is my only option for getting sound back.

Of course if there is a way to get the audio working without throwing
money at a card, I would love to hear it.

While I wait for a reply, I'm going to try getting back to the problem
of the large hard drive and try your suggestions for that. I've been
focusing on the audio issue on the assumption it would be simple and get
resolved quickly to boost my optimism that things were getting better
but that has obviously not gone according to plan. Here's hoping the
hard drive turns out to be easily fixed; I really need a win with this
computer after all the trouble I've been having....

--
Rhino

Paul

unread,

May 17, 2013, 7:18:55 PM5/17/13

to

Find another means to test the motherboard sound ?

You could try a Ubuntu LiveCD, and see if you can get sound
working in there. Sound is every bit as much a PITA in Linux,
as Windows, so really the odds of getting a signal from it,
are about the same. But it's something to try.

In WinXP, I would use setupapi.log to get a bit more info. But
the related files (not the exact same scheme) in Windows 7,
wasn't nearly as helpful.

As Googling isn't digging up any intelligent advice for Code 28
(says to "install drivers"), I'd really have to try the log file
route, and try to gather evidence.

It could be, that InstallShield itself is broken. There is usually
cruft from InstallShield, that stays resident on the computer. I've had to
do something to that, on one occasion.

Another approach, would be to unzip the
VIA_Win7-64_Win7_Vista64_Vista_XP64_XP(v9700a).zip
file, and navigate to the driver folder in there,
right click the appropriate INF, and select "Install"
from the menu. That avoids all the VIA add-on software,
like VDeck or whatever, and just installs a driver. But
the odds of you or I, figuring out which INF to use,
are pretty slim. That's what the main setup code is for.

Paul

Rhino

unread,

May 18, 2013, 9:30:29 AM5/18/13

to

I've got a lot of programs in Add/Remove Programs but I'm not seeing
anything relevant except "Seagate Disk Wizard"; there is no separate
driver entry, just the one for the program itself. There's nothing under
Acronis at all.

However, Device Manager has a category called Acronis Devices which
contains two entries: Acronis Backup Archive Explorer and Acronis
Virtual Disk Bus. Should I remove those??

> Then, go to Disc Wizard, and use the Allocate thing,
> as I explained in my other post, and see if Disk Wizard will reinstall
> the driver. After a reboot, all three partitions should show up.
>

I'm sorry, I'm not finding anything about Allocate in your other posts.
Can you tell me the date and time of the post so that I can locate it?
I've looked at all of the ones since I got the computer back and
restarted the thread but I'm not seeing it.

For what it's worth, I launched Disk Wizard again and it's still
complaining about the exact same sector ("Error: Failed to read from
sector '1,571,841,771' of hard disk 3.") I get options for Retry,
Ignore, Ignore All and Cancel. Retry just gets me the same message
again. I don't want to press Ignore or Ignore All for fear that it will
overwrite something on the disk and then really destroy my data. The
other day, I just clicked the Red X to exit the application but today I
clicked on Cancel and got a slightly different result: I got a sales
pitch for some of their other products and, beneath that, the DiscWizard
application. The main options were Backup, Recovery, Log and Tools &
Utilities. When I clicked on Extended Capacity Manager, I got
"Information: The entire space is accessible: there are no
Seagate/Maxtor hard discs with MBR partition style larger than 2 TB in
your system or sector's size on your disk is larger than 512 KB. All
hard drives can be managed by Add New Disc Tool."

When I went into Add New Disc Wizard, it shows three drives. For some
reason the SECOND one is grayed out. When I click on the first drive: it
says that it is my D: drive and that it is IDE(2) Primary Master. (This
is a 750 GB drive, although only 698.6 GB is actually usable.) I can't
click on the second drive since it is grayed out but it is labelled
IDE(2) Primary Slave and is also 698.6 GB. This second one is clearly my
C: drive although I'm baffled/concerned by why it is grayed out. Maybe
because it is the boot drive and shouldn't be messed with??

Anyway, on to the third disk. It shows a capacity of 746.5 GB which
seems about right; I had split that 3 TB into 3 unevently sized chunks
and 746.5 sounds about right; I think one of them was that size. This
drive is subdivided into TWO areas: one that is my F: drive which is
shown as 746.5 GB NTFS; the other is described as 1.271 TB NTFS and has
a red X in it. I'm not sure if this is my E: drive or my G: drive and I
don't see where the third chunk of space is. The space doesn't add up to
3 TB either. Also, there is a line that says "MBR partition table will
be created on a new disc".

I'm really not sure what all this means and I'm certainly not about to
change anything blindly. If I just start subdividing that space again,
I'll surely lose my E and/or G drives, right?

I"m guessing that the bad sector that DiscWizard detects on startup is
why it is mis-reporting the layout of the 3TB drive; perhaps that is
where some of the partitioning information is stored and it can't read
that sector so it can't tell where the boundary is between two of the
three chunks of the 3TB drive so it only sees two chunks, not three, and
puts the error flag on the bigger chunk to show that something is messed
up there.

Could my mishap with the memory have damaged the large hard drive? I'm
not sure why putting memory into a hot board would mess up one of the
drives but there may be a perfectly logical explanation for that.

At the time, I could barely afford one large drive. My financial
situation is no better now, unfortunately....

> The second disk,
> holds your backup tool images of the contents of the disk.
> If you have a backup, then you never have to worry. Now,
> preparing for the mess you're in here, isn't easy. I don't know
> if there is a way to set up a second 3TB drive, that would not
> be using the services of the Acronis virtual driver. I would
> be looking for some way to avoid using that. Maybe the services
> of two 2TB drives, could provide backups for the three partitions
> on the 3TB drive.
>
> If you had backups, you would not have to worry about trashing
> anything. I don't think there is anything to worry about in
> this case, but what do I know :-(
>

Did you see anything in the way of a support forum like this newsgroup
where I could describe my situation to people who are experts with
DiscWizard? I think you're doing a great job but it sounds as if you are
having to make educated guesses on some points. A true DiscWizard guru
would be a real life saver here as we try to untangle this mess.

--
Rhino

Paul

unread,

May 18, 2013, 10:51:38 AM5/18/13

to

Rhino wrote:

> I've got a lot of programs in Add/Remove Programs but I'm not seeing
> anything relevant except "Seagate Disk Wizard"; there is no separate
> driver entry, just the one for the program itself. There's nothing under
> Acronis at all.
>
> However, Device Manager has a category called Acronis Devices which
> contains two entries: Acronis Backup Archive Explorer and Acronis
> Virtual Disk Bus. Should I remove those??
>
>> Then, go to Disc Wizard, and use the Allocate thing,
>> as I explained in my other post, and see if Disk Wizard will reinstall
>> the driver. After a reboot, all three partitions should show up.
>>
> I'm sorry, I'm not finding anything about Allocate in your other posts.
> Can you tell me the date and time of the post so that I can locate it?
> I've looked at all of the ones since I got the computer back and
> restarted the thread but I'm not seeing it.

In the picture, you can see the "Extended Capacity Manager".

(Click "Apply" to install driver)

http://imageshack.us/a/img805/4338/allocateinstall.gif

I captured that picture, while viewing the video on the
Seagate page. To find it again

http://www.seagate.com/support/downloads/beyond-2tb/

then use the link on the lower right, to see the Flash movie.
The text on the link says:

"Migrate your data to 3TB drive on Windows XP"

I scrolled part way through the video, and took a screen shot.

My guess at this point, is the ability to access that particular sector,
is related to the virtual driver somehow. It's not a failure of the
actual sector. And to figure it out, if that partition is the
second partition, the sector number could be an offset to the
first partition, and when you add the numbers together, the number
is greater than the 2.2TB mark.

Sitting here, I'm having a good deal of trouble just visualizing what
is going on.

I suggested concentrating on getting the driver reinstalled (a filter
driver), in the hope that the request for a particular sector, would
get translated properly by that driver. With the driver missing,
the math that works out the sectors will be wrong. That is how I interpret
the symptoms. And just for the record, I considered this possibility,
when I first read about it. Honestly, my jaw dropped when I read about
how they were going to solve "3TB on older OSes". It's just a monumentally
stupid idea. And what has happened to you, is *exactly why* this is
monumentally stupid.

I never try to partake of storage concepts, for which there are no
recovery tools. Take for example "Dynamic Disk" in Windows. There is
*no* support for that in Linux. I cannot reach for a Linux LiveCD and
rescue myself from Dynamic Disk problems. Consequently, I have to be
careful to purge the Dynamic Disk concept from any of my systems here.

The same kind of thinking goes for the 3TB disk solution. Yes, it's
all "smiles and chuckles" when it works the first time. It's all
"sad faces" later when it is broken.

I don't know if Acronis makes a Linux driver, such that you could
see your data from within Linux. That's another possibility.

Take comfort, that none of your data is lost. Only an attempt to
*repartition* the thing, is going to mess it up. You haven't lost
anything, yet. And no, it's unlikely the memory thing, did anything
to the disk.

Well, the Seagate forum, is going to have unhappy users in it,
reporting problems with DiscWizard. Western Digital, also uses the
rebranded Acronis software, for solving their problems. Those
would be two potential forums to consult. You could also look at the
Acronis forum itself for inspiration. It's a matter of
determining whether the DiscWizard is Acronis TIH or what
equivalent Acronis product it is.

OK, it is " Acronis True Image OEM Version". So it's some
kind of True Image version, baked for the disk companies specially.
It should use the same virtual driver concept as the Acronis True Image.
Perhaps filing a ticket with Acronis tech support, they could suggest
what tool of theirs would make the data accessible again.
Acronis has a Rescue CD, but I doubt this has anything to
do with your problem (it doesn't actually repair anything).

http://kb.acronis.com/content/1836

I still think some of the information in that
Binary._847C629B04FEC51C40D8CCEDBD6BF452 thing I saw in
7ZIP, could be of help. It's the virtual disk driver
installer. And

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Class\{4D36E967-E325-11CE-BFC1-08002BE10318}

is probably missing its LowerFilter entry. That's why I was
trying to use the Extended Capacity Manager (without re-partitioning
anything!), to reload the driver.

Paul

Paul

unread,

May 20, 2013, 12:09:12 AM5/20/13

to

This is a status update, on experiments with Acronis.

In fact, Acronis has been causing some grief for people.

http://forum.acronis.com/forum/31741

Someone here, gives information on what it really takes to
remove the Acronis software. There are two problems. The
extent to which Acronis properly records things in Add/Remove
Programs. And the extent to which their uninstall code,
actually uninstalls things. It does a piss-poor job of
uninstalling.

http://www.wilderssecurity.com/showthread.php?t=316447&page=1

Removal could be considered to be three steps.

1) Find the Acronis product entry in Add/Remove Programs and
uninstall. This leaves a number of relatively benign services.

2) Use ATIH Cleanup.

http://download.acronis.com/ATIH_CleanUp_2012.exe

Now, doing that is largely pointless. On the one hand, it
makes the OS cosmetically cleaner. But it doesn't address
removal of the filter driver that intercepts disk calls.
Touching that can cause the OS to blue screen. The little
cleanup utility, if you look at the INF files, is a
cleanup in name only.

3) The Wilderssecurity thread, is an attempt by someone who
visited the Acronis forums, to help, got booted from the
forum, and sought to post the info elsewhere, to help end
users. Unfortunately, the info isn't in completely friendly
"sevenforums.com" style recipe. While I have the backups
needed of my system, to attempt to do this, at the moment
I have no need to explore this. I didn't try to clean
any Acronis stuff from my system.

*******

OK, so why is this important ?

I decided to try to reproduce a data recovery effort on a 3TB
drive, using Acronis as a "naive customer". I was trying to
reproduce your problem.

It turns out, due to the un-hygienic nature of the Acronis approach,
I *already had* the Acronis filter driver in place! It appears
I must have installed, for evaluation, the Seagate DiscWizard at
some point in the past. I don't remember doing it, but the
fact I had the accursed filter driver already in my OS, proves
it.

So what happened this time ? This is what I tried.

1) Install 3TB disk.
2) Install Acronis TI OEM (a.k.a Western Digital tool for large disks)
This is effectively the same software as Seagate DiscWizard.
3) I can access the bottom 2.2TB. I cannot access the top 0.8TB.
No way, no how. It won't let me. And the old filter driver it
cannot remove or update, is the reason why!

So then the fun started. Before I began, I did a backup of my system.
I had a backup from a year ago, which I restored, and the Acronis
filter driver was in that one. So that was polluted. I had to go
back two years. Lucky I still had my 06-2011 backup image on
one of the disks in my spare disk pile.

I installed that, checked, and no Acronis filter driver. I
installed the Acronis OEM software, went to the Extended Capacity
Manager, and it worked! I could access the upper portion of the
disk, as two disks. The physical disk, and a virtual disk. And
these are all done, by the filter driver stuff.

Now, if you actually partition inside the Acronis tool, the Acronis
tool does partitions suitable for Windows 7, rather than WinXP.
Partition offset is "1 megabyte", suited to SSDs perhaps. If
I do the partitioning in WinXP itself, the offset is 63 sectors.

So anyway, this is the setup.

+------+----------------+-----------------------+
| MBR | 62 sectors gap | 2.2TB NTFS partition |
+------+----------------+-----------------------+

+------------------+------+-------------------+-------------------------+
| 262144 byte gap | MBR | ~1048576 byte gap | 746.5GB NTFS partition |
+------------------+------+-------------------+-------------------------+

That means, I actually used Windows Disk Management partitioning, for
the immediately visible 2.2TB portion. And used the Acronis Extended
Capacity Manager window, to place a partition on the second chunk.
I suppose I could have gone back, in Disk Management, and deleted and
re-made the partition for the 746.5GB chunk, but I decided to just
leave it. The first partition, is legacy DOS compatible "63 sector"
geometry, while the second is more in tune with Windows 7 "1 megabyte"
offset.

********

OK, so now I have a two-year old version of my system, with a
successful partitioning. I put a couple test files in the upper
portion, as a test.

Now, the question was, what would happen, if I lost access to the
upper portion. For example, if I were to go forward, to my 2013
operating system image, I'd lose access to the upper part.

It turns out, I was able to do data recovery for the upper portion
from Linux. You can do what I did, as long as there is storage
space on the lower partition, to hold a bitmap copy of the thing
labeled "746.5GB" in the above diagram. By using the "disk dump"
program known as "dd", I could make a copy of the upper partition,
then loopback mount it for later.

*******

If you don't want to install the entire Acronis TI OEM, you can just
do the driver portion.

(1,890,091 bytes)

http://kb.acronis.com/system/files/content/2013/01/38937/virtualdisksetup.zip

Unzipped, gives VirtualDiskSetup.msi .

I was able, by installing that on the clean system from 2011, to
regain access to both partitions. So that's another tool option,
if you wanted to try it (like, just after a clean system install
perhaps).

Note that it doesn't create an entry in Add/Remove, and
doesn't clean up after itself etc.

*******

Now, your situation, I think, is something like this.

+------+-----+-----------------------+------------------------+
| MBR | gap | 746GB NTFS partition | 1454GB NTFS partition |
+------+-----+-----------------------+------------------------+

+------+------+-------+-----------------------+
| gap | MBR | gap | 746GB NTFS partition |
+------+------+-------+-----------------------+

And I think you're telling me, the 1454GB one, has gone missing.
You cannot see it.

My guess is, this is related to the NVidia Southbridge, its RAID
capability, and the inclusion of space to store metadata, at the end
of the disk. The NVidia driver (MediaShield), probably wants to leave
space at the end, to store metadata, the day you decide to switch.

Now, perhaps you left the BIOS setting for the NVidia ports at
"RAID", then did JBOD (single disk installation). The NVidia driver
then makes the disk "a couple inches shorter" than it would normally
be. I'm going to redraw that as follows. NVidia makes room for
its metadata, at the peril of snipping off the end of your partition!
This happens, because normally you partition *after* the driver
is installed, and then your partition would be the proper length.

+------+-----+-----------------------+------------------------+--------+
| MBR | gap | 746GB NTFS partition | 1453GB NTFS partition |metadata|
+------+-----+-----------------------+------------------------+--------+

There is a lesson to be learned here. *Never* let the last partition,
touch the "right hand edge" of the disk! Leave a gap on the end, for
stupid stuff like this.

OK, now the info in the NTFS file system, says it is 1454GB long.
NVidia has snipped the end off the partition. The "cannot access block 15xxxxxx"
thing, is caused by an attempt to reach the end of the partition, an
end which is no longer accessible. You are requesting read access,
to a block which is past the end of the disk (as far as the NVidia
*driver* is concerned, not the disk drive itself - the driver is
responsible for messing this up).

Moving the disk to another controller, might change this.

Changing the setting in the BIOS might change it.

Your guy at the shop, would have to make a decision, while
installing the new motherboard, as to what mode to run
in. On page 44 of the N68C-GS FX PDF manual, I see

Onboard SATA Controller [Enabled]
SATA Operation Mode [IDE]

and I feel that would cause the least trouble. If I did that
on an Intel chipset, I could probably get the thing to boot,
even though there was no driver for the chipset, when the new
motherboard was put into place. I don't know exactly, how
I'd get the RAID driver in there, unless perhaps I did a
"repair install" of WinXP.

So anyway, your second partition might reappear, if:

1) Fresh OS (no Acronis).
2) SATA port with no need for a metadata section (IDE mode).
3) At this point, you have access to the first 2.2TB of storage.
4) Install VirtualDiskSetup.msi or use Acronis TI OEM (which
installs the same files), to put the filter driver in place.
Reboot. The upper (virtual) partition will now be visible.
All three partitions should work.

*******

Now, back to the data recovery tutorial.

The problem is, many disk access methods, won't let you go above
the 2.2TB mark. If you wanted to rescue the upper 746.5GB partition,
you could find a way of moving it down, and storing the
entire partition, as a "file". The "file" can be mounted in
Linux, using a loopback mount. You need sufficient space to
do this.

Since my bottom 2.2TB partition is empty, I can easily store
a single 800GB image file in it. Then, I can do a file "copy and
paste", from that mounted image, and store the 800GB of loose
files in the 2.2TB partition.

This is how I developed my picture of how the partitions
are laid out, by the Acronis stuff.

1) Use Windows tools, to determine the sizes of things. There
are several potential sizing tools. Everest Home Edition,
PTEDIT32 (for viewing the first MBR), and the port of "dd".

This is the info I recorded in my crib notes. I gathered
this information, while the thing was still working. And I
did that, to help me decode and verify things. The first
row, is the size of the disk, while the second line
is the size of the file system. Converting the second
numbers to hexidecimal was crucial, for reading the second
MBR by hand. It turned out, the bytes were in reverse order,
in my hex viewer.

801,569,243,136 = 0x5D50A000
2199020350464 (63???) 801,568,194,560 (1meg???) = 0x5D509800

2) Boot Ubuntu. I keep a pendrive with a recent Ubuntu on it.
A version of Ubuntu without Unity on it. Just nice regular
menus for the programs.

The 3TB drive, ends up on my system as /dev/sdd. Linux
does not "see" the upper 746.5GB partition, because there
is no Acronis filter driver for it.

3) The hexdump program, allows me to read the disk sectors.
In these two example commands, I was looking for the two
MBR sectors on the disk. The number 0x20000000000 hex,
is the 2.2TB thing. The number 40000 hex, is the offset
from that boundary, until I found the MBR. When I view
these on my Linux screen, I can see the MBR boot message
text, which is identical (it would be the boot sector
code that Acronis put there). Since I don't boot from this
disk, there is no Windows boot code. The last bytes of an
MBR has 0xAA55 in it, which is an additional check you've
found the MBR. So, the first command, uses no offset, and
reads the physical MBR at sector 0 offset. The second
command, goes to "2.2TB + 262144 offset", where I found
the MBR. I had to search a tiny bit, to find it.

sudo hexdump -C -n 512 /dev/sdd <--- MBR

sudo hexdump -C -s 0x20000040000 -n 512 /dev/sdd <--- MBR

While looking at the short dump from the second command, I see:

200000401b0 00 00 00 00 00 00 00 00 d1 1e 03 02 00 00 80 20 |............... |
200000401c0 21 00 07 fe ff ff 00 08 00 00 00 98 50 5d 00 00 |!...........P]..|

Offset 0x1BE, is where the 16 bytes of the partition definition
are stored. And 0x1BE + C = 0x1CA, is where the size of the
partition is stored. That's the "00 98 50 5d" part. Now, if you
glance up the screen, you can see the value is actually
0x5D509800 bytes or 801,568,194,560 bytes. That's the amount
of information I must capture from the top end of the disk,
to store an image of the partition in the lower 2.2TB blank
space. By comparing the numbers I was expecting to see,
with what showed up in the hex editor, I was able to figure
out I needed to reverse the byte order. So I'm pretty sure
I got the whole partition.

The next step, is verifying I'm looking at the NTFS headers.
The letters NTFS, fortunately appear in each one. The commands
I used to look at them were.

sudo hexdump -C -s 32256 -n 512 /dev/sdd <--- first partition

sudo hexdump -C -s 0x20000140000 -n 512 /dev/sdd <--- second partition

The 32256 decimal number, is 63*512 bytes or 63 sectors. Since
Disk Management prepared that partition, the offset is 63
sectors.

The 0x140000 hex offset, is the Acronis partitioning, and
Acronis offset leaving room for their metadata. I presume
somewhere in that 0x40000 section, Acronis stores some
info for their filter driver. I did not verify that.
The additional 0x100000 is the 1megabyte offset that
Windows 7 likes.

So both of those hexdump screens, show me an NTFS header.
The info I have in hand now, is I know the offset where the
second NTFS partition starts, I also know its length. Now,
it's time to transfer the entire upper partition, as a file.

Steps to transfer.

Using the Linux "factor" program, figure out what
block size might be used.

factor 801568194560 ==> 2**20 5 7 21841

I decide to use 262144 as a block size, as the offset
is measured in blocks rather than bytes. And I need
to ensure I can skip to the right location to start the
transfer.

Length is 801568194560 / 262144 = 3057740 blocks.

Skip 0x20000140000 but in blocks. If I convert
that number to decimal, using the Linux "calculator",
then divide by 262144, I get a skip of 8388613 blocks.

In Windows, my 2.2TB partition was called "GUMBY", and
the 746.5GB partition was called "POKEY". We're storing
an image of POKEY, in the NTFS GUMBY file system. This
is the command that does the transfer, at 120MB/sec. It
took around two hours, to complete the imaging of the
upper partition.

bs=262144 The block size I chose for "dd" program
skip=8388613 Start transferring from 0x20000140000 byte.
Converted the number to "blocks"
count=3057740 801568194560 / 262144 = number of blocks
to transfer.

The "dd" command is piped to the GNU Linux "cp" or copy
program. The "/proc/self/fd/0" means the same thing as
"stdin" - it's a way to fool "cp" to use the byte stream
from the "dd" program, as the "input file". The "--sparse"
option, says "for every sector which is completely zeros,
don't store the result on disk". This is a way of
making a sparse file. Since the disk is new, and the partitions
are mostly full of zeros, this saves a *lot* of space, and
makes the transfer operation run 4X faster (as a lot less
writes are needed).

sudo dd if=/dev/sdd bs=262144 skip=8388613 count=3057740 | cp --sparse=always /proc/self/fd/0 /media/GUMBY/pokey.dd

The actual output file is /media/GUMBY/pokey.dd .

If I view that file in Windows at the moment, it reports:

http://imageshack.us/a/img818/7466/pokeysize.gif

Size: 746GB

Size on disk: 216MB

Since any zeroed portions of the partition are not written
out in this case, the operation used very little of the
lower 2.2TB partition. This only works, if you know the
partition is mostly zeroed.

http://en.wikipedia.org/wiki/Sparse_file

If you wanted no sparse file to be created, then the "dd"
command would have looked like this. I started doing this,
but stopped it after a couple hours, because it was just
taking too long. This would take exactly 746GB of storage
space, as no zeroed sectors are removed.

sudo dd if=/dev/sdd of=/media/GUMBY/pokey.dd bs=262144 skip=8388613 count=3057740

Now, in Linux, I make a mount point for my new bitmap image.

sudo mkdir /media/POKEY
sudo mount -t ntfs -o loop /media/GUMBY/pokey.dd /media/POKEY

Magically, the POKEY window opens, and I can do a copy and
paste file transfer, to start recovering the files from
POKEY partition. The 746GB file is now mounted as a file
system. I don't want to attempt to write to it, as
that would not be very efficient. But it's still
fully functional.

So that's how I can do data recovery, on an Acronis
setup, where the top partition is no longer visible.

This doesn't help you, because I suspect the end of your
1454GB partition, has had a few sectors snipped off the
end, by the NVidia RAID-ready MediaShield driver.

+------+-----+-----------------------+------------------------+--------+
| MBR | gap | 746GB NTFS partition | 1453GB NTFS partition |metadata|
+------+-----+-----------------------+------------------------+--------+

It's possible, that by switching over to Linux, the
1454GB would be visible there, and you could do
data recovery without any nonsense at all. That
would be something to investigate. Just boot
into Linux and have a look.

It's only if you want to do data recovery *above* 2.2TB,
that the "dd" command can oblige. Being a block level
command, it does not know or care about the 32 bit
limit on sector addresses. That's how I can roam above
2.2TB and grab a bitmap copy of the partition up
there.

And the above techniques can all be done on the same
disk, if there is space. If your disk was chock
full, this approach would be a non-starter. To do data
recover on a chock-full disk, you really need to buy
another disk.

HTH,
Paul

Rhino

unread,

May 21, 2013, 2:20:20 PM5/21/13

to

From your remarks, I'm not very optimistic about Ubuntu but I've
started a download of Ubuntu 12.04 LTS. I'm finding the web pages rather
confusing though; I'm not clear on whether the downloaded file is going
to be a "LiveCD" (which appears to be a version of Ubuntu that I can run
from the CD without deleting or hurting Windows in any way)
automatically or if I have to do something to make sure that it is
treated as a LiveCD and not an install CD. I REALLY don't want to
clobber Windows at this stage; that would really be pouring salt in the
wounds.

> In WinXP, I would use setupapi.log to get a bit more info. But
> the related files (not the exact same scheme) in Windows 7,
> wasn't nearly as helpful.
>

I did a search on my whole C: drive (assuming it would be there) for
setupapi.log. I found two similarily named files - setupapi.log.0.old
and setupapi.log.1.old - in C:\Windows but NOT setupapi.log itself. I'm
not sure where it is. I'm guessing that as an active log, it is only
accessible to the processes that right that log. I'm guessing that I
need to do something specific to change logs so that the current one
becomes an old one but have no idea what I need to do.

For what it's worth, I opened the setupapi.log.1.old file and found that
it covered the period 2012/01/18 through 2013/05/17. I got the computer
back from the tech on 2013/05/11 so this file actually has at least some
of what you likely want to see. The problem is that it is huge (1,923
KB) and newsgroups don't offer an easy way to do attachments.
Unfortunately, the log is relatively cryptic and I can't just see at a
glance what is wrong with my desktop and why it so stubbornly refuses to
install my audio drivers.

Do you know of some public place where I could upload that log for you
(and any others) could view it to help me with my problem? Otherwise, I
can set up a free hosting account somewhere, upload to there, and give
you the URL.

> As Googling isn't digging up any intelligent advice for Code 28
> (says to "install drivers"), I'd really have to try the log file
> route, and try to gather evidence.
>
> It could be, that InstallShield itself is broken. There is usually
> cruft from InstallShield, that stays resident on the computer. I've had to
> do something to that, on one occasion.
>
> Another approach, would be to unzip the
> VIA_Win7-64_Win7_Vista64_Vista_XP64_XP(v9700a).zip
> file, and navigate to the driver folder in there,
> right click the appropriate INF, and select "Install"
> from the menu. That avoids all the VIA add-on software,
> like VDeck or whatever, and just installs a driver. But
> the odds of you or I, figuring out which INF to use,
> are pretty slim. That's what the main setup code is for.
>

I was going to offer to just try clicking on all the INFs in turn to see
if one of them fixed my audio. Unfortunately, there are 52 INFs in that
zip file. If each needs a reboot to be sure it works, I'd be at it for
hours.... Can we narrow down which ones to try/avoid? If I can get that
to a substantially smaller number, that might be the most direct way to
sort this mess out.

Hmm. Now that I've expanded the width of the "In Folder" column, I think
that only the first six INFs, the ones in the HDUAA folder, are viable
candidates. All the rest are for Versions of Windows after XP or 64-bit.
Since six is a lot more palatable than 52, I'm going to try them. I
right-clicked on the first one, chose Install and then... nothing
happened. Shouldn't I get some kind of install dialog??

I tried doing the install again in Device Manager (clicking on the
"unknown device" entry in Sound, video and game controllers) and, once
again, the install failed with error code 28. However, I still had the
search window open for setupapi.log and, suddenly, a very short file
named (exactly) setupapi.log appeared. Based on the timestamp, it
apparently covers that latest attempt to install the driver. I'm
including a paste of the entire file here:
========================================================================
[SetupAPI Log]
OS Version = 5.1.2600 Service Pack 3
Platform ID = 2 (NT)
Service Pack = 3.0
Suite = 0x0100
Product Type = 1
Architecture = x86
[2013/05/21 13:59:51 4860.628]
#-198 Command line processed: "C:\WINDOWS\system32\mmc.exe"
C:\WINDOWS\system32\devmgmt.msc /s
#I060 Set selected driver.
#-019 Searching for hardware ID(s):
hdaudio\func_01&ven_1106&dev_4397&subsys_18490397&rev_1000,hdaudio\func_01&ven_1106&dev_4397&subsys_18490397
#-018 Searching for compatible ID(s):
hdaudio\func_01&ven_1106&dev_4397&rev_1000,hdaudio\func_01&ven_1106&dev_4397,hdaudio\func_01&ven_1106,hdaudio\func_01
#I393 Modified INF cache "C:\WINDOWS\inf\INFCACHE.1".
#I022 Found "HDAUDIO\FUNC_01&VEN_1106&DEV_4397&SUBSYS_18490397" in
C:\WINDOWS\inf\oem0.inf; Device: "VIA High Definition Audio"; Driver:
"VIA High Definition Audio"; Provider: "VIA Technologies, Inc."; Mfg:
"VIA Technologies, Inc."; Section name: "HdAudModel".
#I023 Actual install section: [HdAudModel]. Rank: 0x00000001. Effective
driver date: 06/14/2011.
#I022 Found "HDAUDIO\FUNC_01&VEN_1106&DEV_4397" in
C:\WINDOWS\inf\oem68.inf; Device: "VIA High Definition Audio"; Driver:
"VIA High Definition Audio"; Provider: "VIA Technologies, Inc."; Mfg:
"VIA Technologies, Inc."; Section name: "HdAudModel".
#I023 Actual install section: [HdAudModel]. Rank: 0x00002001. Effective
driver date: 06/14/2011.
#-019 Searching for hardware ID(s):
hdaudio\func_01&ven_1106&dev_4397&subsys_18490397&rev_1000,hdaudio\func_01&ven_1106&dev_4397&subsys_18490397
#-018 Searching for compatible ID(s):
hdaudio\func_01&ven_1106&dev_4397&rev_1000,hdaudio\func_01&ven_1106&dev_4397,hdaudio\func_01&ven_1106,hdaudio\func_01
#I022 Found "HDAUDIO\FUNC_01&VEN_1106&DEV_4397" in
d:\downloads\via_win7-64_win7_vista64_vista_xp64_xp(v9700a)\audio(v9700a)\viahdaud\present\drivers\hduaa\viahdb.inf;
Device: "VIA High Definition Audio"; Driver: "VIA High Definition
Audio"; Provider: "VIA Technologies, Inc."; Mfg: "VIA Technologies,
Inc."; Section name: "HdAudModel".
#I023 Actual install section: [HdAudModel]. Rank: 0x00002001. Effective
driver date: 06/14/2011.
#I022 Found "HDAUDIO\FUNC_01&VEN_1106&DEV_4397" in
d:\downloads\via_win7-64_win7_vista64_vista_xp64_xp(v9700a)\audio(v9700a)\viahdaud\present\drivers\hduaa\viahdb1.inf;
Device: "VIA High Definition Audio"; Driver: "VIA High Definition
Audio"; Provider: "VIA Technologies, Inc."; Mfg: "VIA Technologies,
Inc."; Section name: "HdAudModel".
#I023 Actual install section: [HdAudModel]. Rank: 0x00002001. Effective
driver date: 06/14/2011.
#I022 Found "HDAUDIO\FUNC_01&VEN_1106&DEV_4397&SUBSYS_18490397" in
d:\downloads\via_win7-64_win7_vista64_vista_xp64_xp(v9700a)\audio(v9700a)\viahdaud\present\drivers\hduaa\viahdb2.inf;
Device: "VIA High Definition Audio"; Driver: "VIA High Definition
Audio"; Provider: "VIA Technologies, Inc."; Mfg: "VIA Technologies,
Inc."; Section name: "HdAudModel".
#I023 Actual install section: [HdAudModel]. Rank: 0x00000001. Effective
driver date: 06/14/2011.
#-166 Device install function: DIF_SELECTBESTCOMPATDRV.
#I063 Selected driver installs from section [HdAudModel] in
"c:\windows\inf\oem0.inf".
#I320 Class GUID of device remains: {4D36E96C-E325-11CE-BFC1-08002BE10318}.
#I060 Set selected driver.
#I058 Selected best compatible driver.
#-166 Device install function: DIF_SELECTBESTCOMPATDRV.
#I063 Selected driver installs from section [HdAudModel] in
"c:\windows\inf\oem0.inf".
#I320 Class GUID of device remains: {4D36E96C-E325-11CE-BFC1-08002BE10318}.
#I060 Set selected driver.
#I058 Selected best compatible driver.
#-124 Doing copy-only install of
"HDAUDIO\FUNC_01&VEN_1106&DEV_4397&SUBSYS_18490397&REV_1000\4&1F7F6F18&2&0001".
#-166 Device install function: DIF_REGISTER_COINSTALLERS.
#I056 Coinstallers registered.
#-166 Device install function: DIF_INSTALLINTERFACES.
#-011 Installing section [HdAudModel.Interfaces] from
"c:\windows\inf\oem0.inf".
#I054 Interfaces installed.
#-166 Device install function: DIF_INSTALLDEVICE.
#I123 Doing full install of
"HDAUDIO\FUNC_01&VEN_1106&DEV_4397&SUBSYS_18490397&REV_1000\4&1F7F6F18&2&0001".
#-035 Processing service Add/Delete section [HdAudModel.Services].
#E279 Add Service: Failed to create service "KaraokeService". Error
1078: The name is already in use as either a service name or a service
display name.
#E033 Error 1078: The name is already in use as either a service name or
a service display name.
#E275 Error while installing services. Error 1078: The name is already

in use as either a service name or a service display name.

#E122 Device install failed. Error 1078: The name is already in use as

either a service name or a service display name.

#E154 Class installer failed. Error 1078: The name is already in use as

either a service name or a service display name.

#I060 Set selected driver.
#I125 Installing NULL driver for
"HDAUDIO\FUNC_01&VEN_1106&DEV_4397&SUBSYS_18490397&REV_1000\4&1F7F6F18&2&0001".
#I121 Device install of
"HDAUDIO\FUNC_01&VEN_1106&DEV_4397&SUBSYS_18490397&REV_1000\4&1F7F6F18&2&0001"
finished successfully.
=========================================================================

Apparently, the KaraokeService is what's giving us trouble here. I never
do karaoke in bars and haven't consciously installed anything intended
to help me karaoke at home so if this is just a matter of uninstalling
something karaoke-related, I'd be fine with that. But for all I know,
that's just their term for a whole slew of audio services, some of which
I _do_ need.

[Sorry for the delay in getting back to this. I just can't face this
stuff some days, especially when I'm not making an inch of headway.]

--
Rhino

Rhino

unread,

May 23, 2013, 6:19:27 PM5/23/13

to

On 2013-05-20 00:09, Paul wrote:
> This is a status update, on experiments with Acronis.
>

First of all, Paul, I want to thank you again for making such an
outstanding and time-consuming effort to help me with this. You've gone
way above and beyond the call of duty in your efforts on my behalf.

Frankly, I skimmed this a couple of days ago and found it very
intimidating. There is a lot of information here and I'm frankly not
following all of it. I procrastinated for a couple of days then I came
back to read this all again in detail.

I have no full backup of the data on the 3 TB drive. I've actually spend
part of the last couple of days looking over my DVD-RW disks and finding
a small percentage of the files but nowhere near all of them. And that's
all I have in the way of backups. I've always known that there's a
chance I'd simply lose everything on that drive and really regretted the
thought but couldn't afford to prevent it by buying a second hard drive
just to serve as backup. (Considering it would also have been 3 TB in
size, I expect the backup drive would have had the same problem after
the mobo swap anyway! Then again, I suppose the second drive should have
been in an external enclosure and that it would then not have been
subject to the same problem. Or would it? I'm not sure what happens with
respect to OS to an external drive.)

I'm not adverse to poking around in the registry and manually deleting
programs, although I'd want a very clear and detailed list of EXACTLY
what needed to be removed and HOW it is to be done. The wildersecurity
thread you cite may be as close as exists and if that is the case, it's
not close enough. I don't know what Arvy means when he says "remove only
inserted values" (from the registry) nor how do do that correctly; I
don't know exactly which required to boot services he wants removed; I
don't know how which specific residual Acronis services, drivers files
and miscellaneous leftovers he wants removed. He gives a rather detailed
list of things to do above his summation but it appears that my version
of the Acronis software may not have exactly the same things as his so I
could easily fail to remove something vital simply because it is not in
his list. Just following his instructions as closely as possible and
hoping for the best doesn't seem like a formula for success. I'm bound
to mess SOMETHING up.

So, if I could cleanly remove all of DiscWizard and the Acronis drivers,
then install the software again, I'd probably be fine. But that's a
mighty big IF!!

> Now, if you actually partition inside the Acronis tool, the Acronis
> tool does partitions suitable for Windows 7, rather than WinXP.
> Partition offset is "1 megabyte", suited to SSDs perhaps. If
> I do the partitioning in WinXP itself, the offset is 63 sectors.
>
> So anyway, this is the setup.
>
> +------+----------------+-----------------------+
> | MBR | 62 sectors gap | 2.2TB NTFS partition |
> +------+----------------+-----------------------+
>
> +------------------+------+-------------------+-------------------------+
> | 262144 byte gap | MBR | ~1048576 byte gap | 746.5GB NTFS partition |
> +------------------+------+-------------------+-------------------------+
>
> That means, I actually used Windows Disk Management partitioning, for
> the immediately visible 2.2TB portion. And used the Acronis Extended
> Capacity Manager window, to place a partition on the second chunk.
> I suppose I could have gone back, in Disk Management, and deleted and
> re-made the partition for the 746.5GB chunk, but I decided to just
> leave it. The first partition, is legacy DOS compatible "63 sector"
> geometry, while the second is more in tune with Windows 7 "1 megabyte"
> offset.
>

Sorry, the last couple of paragraphs are going over my head. The
partition size stuff is irrelevant right? I'd obviously want to use the
Windows Disk Management partitioning, right? After all, this is an XP
machine, not a Win7 machine.

> ********
>
> OK, so now I have a two-year old version of my system, with a
> successful partitioning. I put a couple test files in the upper
> portion, as a test.
>
> Now, the question was, what would happen, if I lost access to the
> upper portion. For example, if I were to go forward, to my 2013
> operating system image, I'd lose access to the upper part.
>
> It turns out, I was able to do data recovery for the upper portion
> from Linux. You can do what I did, as long as there is storage
> space on the lower partition, to hold a bitmap copy of the thing
> labeled "746.5GB" in the above diagram. By using the "disk dump"
> program known as "dd", I could make a copy of the upper partition,
> then loopback mount it for later.
>

I know what a loopback address is but have no idea what a loopback mount
is.

> *******
>
> If you don't want to install the entire Acronis TI OEM, you can just
> do the driver portion.
>
> (1,890,091 bytes)
>
> http://kb.acronis.com/system/files/content/2013/01/38937/virtualdisksetup.zip
>
>
> Unzipped, gives VirtualDiskSetup.msi .
>
> I was able, by installing that on the clean system from 2011, to
> regain access to both partitions. So that's another tool option,
> if you wanted to try it (like, just after a clean system install
> perhaps).
>
> Note that it doesn't create an entry in Add/Remove, and
> doesn't clean up after itself etc.
>

So I'd still want to do the Add/Remove and assorted cleanup - once we
know EXACTLY what needs to be done - before trying to use
VirtualDiskSetup.msi, right?

> *******
>
> Now, your situation, I think, is something like this.
>
> +------+-----+-----------------------+------------------------+
> | MBR | gap | 746GB NTFS partition | 1454GB NTFS partition |
> +------+-----+-----------------------+------------------------+
>
> +------+------+-------+-----------------------+
> | gap | MBR | gap | 746GB NTFS partition |
> +------+------+-------+-----------------------+
>
> And I think you're telling me, the 1454GB one, has gone missing.
> You cannot see it.
>

Actually, I've got THREE large chunks. I don't remember the exact
proportions of the original partitions but I can see one of them (my F:
drive) as having size 746.5 GB NTFS; then the remaining two partitions,
which must have totalled 2.2 TB or thereabouts, should be the other two
partitions. But I'm only seeing the 746.5 GB and a SINGLE partition of
1.271 TB NTFS; it has a small X circled in red in the graphical
representation of the space.

> My guess is, this is related to the NVidia Southbridge, its RAID
> capability, and the inclusion of space to store metadata, at the end
> of the disk. The NVidia driver (MediaShield), probably wants to leave
> space at the end, to store metadata, the day you decide to switch.
>
> Now, perhaps you left the BIOS setting for the NVidia ports at
> "RAID", then did JBOD (single disk installation). The NVidia driver
> then makes the disk "a couple inches shorter" than it would normally
> be. I'm going to redraw that as follows. NVidia makes room for
> its metadata, at the peril of snipping off the end of your partition!
> This happens, because normally you partition *after* the driver
> is installed, and then your partition would be the proper length.
>
> +------+-----+-----------------------+------------------------+--------+
> | MBR | gap | 746GB NTFS partition | 1453GB NTFS partition |metadata|
> +------+-----+-----------------------+------------------------+--------+
>
> There is a lesson to be learned here. *Never* let the last partition,
> touch the "right hand edge" of the disk! Leave a gap on the end, for
> stupid stuff like this.
>

One of those things I wished I'd known back when I installed the drive.
But, of course, the horse has long fled the barn so there's no point in
closing the door now. (Aside from preventing further escapes from the
barn in future of course.)

> OK, now the info in the NTFS file system, says it is 1454GB long.
> NVidia has snipped the end off the partition. The "cannot access block
> 15xxxxxx"
> thing, is caused by an attempt to reach the end of the partition, an
> end which is no longer accessible. You are requesting read access,
> to a block which is past the end of the disk (as far as the NVidia
> *driver* is concerned, not the disk drive itself - the driver is
> responsible for messing this up).
>
> Moving the disk to another controller, might change this.
>

How would I do that? Just swap the SATA connectors between my drives?
Would something that simple solve my entire problem? Please, PLEASE tell
me that would be the case! I'd gladly shutdown, power off, and then try
all the different combinations of SATA cables to SATA drives....

> Changing the setting in the BIOS might change it.
>

Which setting? Change it from what to what?

> Your guy at the shop, would have to make a decision, while
> installing the new motherboard, as to what mode to run
> in. On page 44 of the N68C-GS FX PDF manual, I see
>
> Onboard SATA Controller [Enabled]
> SATA Operation Mode [IDE]
>
> and I feel that would cause the least trouble. If I did that
> on an Intel chipset, I could probably get the thing to boot,
> even though there was no driver for the chipset, when the new
> motherboard was put into place. I don't know exactly, how
> I'd get the RAID driver in there, unless perhaps I did a
> "repair install" of WinXP.

I can ask him what he did but at this point, it's nearly two weeks and
he likely won't remember. He said he made some notes but I don't think
they were very detailed.

>
> So anyway, your second partition might reappear, if:
>
> 1) Fresh OS (no Acronis).

You're just talking about removing ALL of Acronis right?

> 2) SATA port with no need for a metadata section (IDE mode).

I'm not following this suggestion at all.

> 3) At this point, you have access to the first 2.2TB of storage.

No, I can SEE one of the partitions in DiscWizard but I can't see any of
the data. If I click on my E: drive, Windows offers to format it.

There is a DiskClone feature in DiscWizard which looks like it would be
willing to make a copy of the 3 TB disk. I'm not sure what kind of mess
I'd get if I actually tried to make the copy with only one of the three
partitions properly identifid in DiscWizard. And I have no idea where
I'd WRITE the copy given that I don't have a spare 3 TB disk. Or are you
proposing that I buy one? Or just buy 3 TB worth of blank DVD-RWs and
start copying?

> 4) Install VirtualDiskSetup.msi or use Acronis TI OEM (which
> installs the same files), to put the filter driver in place.
> Reboot. The upper (virtual) partition will now be visible.
> All three partitions should work.
>

That point sounds fine IF we can resolve all the OTHER issues I've
raised along the way.

> *******
>
> Now, back to the data recovery tutorial.
>
> The problem is, many disk access methods, won't let you go above
> the 2.2TB mark. If you wanted to rescue the upper 746.5GB partition,
> you could find a way of moving it down, and storing the
> entire partition, as a "file". The "file" can be mounted in
> Linux, using a loopback mount. You need sufficient space to
> do this.
>

For what it's worth, I finally got a copy of Ubuntu 12.04 LTS (32-bit)
on a single CD-R. I haven't tried it yet but would be willing to do so.

> Since my bottom 2.2TB partition is empty, I can easily store
> a single 800GB image file in it. Then, I can do a file "copy and
> paste", from that mounted image, and store the 800GB of loose
> files in the 2.2TB partition.
>
> This is how I developed my picture of how the partitions
> are laid out, by the Acronis stuff.
>
> 1) Use Windows tools, to determine the sizes of things. There
> are several potential sizing tools. Everest Home Edition,
> PTEDIT32 (for viewing the first MBR), and the port of "dd".
>

I have Everest Home Edition. It's ancient (Version 2.20.405) but it was
the newest one I could find a few years back when I was looking for it.

I've never heard of PTEDIT32 but I just acquired it. It has the
following information for my 3 TB drive:
====================================================================
Drive 3 (764428 MB) - 97451 cyl, 255 heads, 63 sectors per track

Partitition Table at sector 0 (cyl 0, head 0, sector 1):

STARTING ENDING SECTORS
TYPE BOOT CYL HEAD SECTOR CYL HEAD SECTOR BEFORE SECTORS
1 05 80 1 0 1 1023 254 63 16065 1565534250
2 07 00 1023 254 63 1023 254 63 156555031 2729411370
3 00 00 0 0 0 0 0 0 0 0
4 00 00 0 0 0 0 0 0 0 0
======================================================================

When I click the "Goto EPBR" button, the display changes to this:

====================================================================
Drive 3 (764428 MB) - 97451 cyl, 255 heads, 63 sectors per track

Partitition Table at sector 16065 (cyl 1, head 0, sector 1):

STARTING ENDING SECTORS
TYPE BOOT CYL HEAD SECTOR CYL HEAD SECTOR BEFORE SECTORS
1 07 00 1 1 1 1023 254 63 63 1565534187
2 00 00 0 0 0 0 0 0 0 0
3 00 00 0 0 0 0 0 0 0 0
4 00 00 0 0 0 0 0 0 0 0
======================================================================

I don't know if it helps but I may also have a rather old version of
Partition Magic (Version 4.0). Maybe! (I have the box and the manuals
but I don't see the actual CD. But if I looked around I might be able to
dig it up.) Given how old it is - I think it predates XP by a couple of
years and TB drives were probably not allowed for in software that old -
it's probably useless but I thought I should mention it just in case it
has some conceivable use.

I could certainly try this. But how do I get at the data if I can see
it? Would I be doing this to create a backup of the 3 TB drive? Even
assuming I do this and fill the requisite number of DVD-RWs, what would
happen with my current 3 TB drive after I've done all of this? Would
that just make it safer to try things to fix the drive since I would
know my data was safe but still risk losing all the data on the original
drive?

> It's only if you want to do data recovery *above* 2.2TB,
> that the "dd" command can oblige. Being a block level
> command, it does not know or care about the 32 bit
> limit on sector addresses. That's how I can roam above
> 2.2TB and grab a bitmap copy of the partition up
> there.
>
> And the above techniques can all be done on the same
> disk, if there is space. If your disk was chock
> full, this approach would be a non-starter. To do data
> recover on a chock-full disk, you really need to buy
> another disk.
>

I hadn't looked at the disk-available/disk-used numbers in a few months
and don't remember them clearly in any case. I know the three partitions
were definitely NOT full but each had a fair percentage of data on it,
probably more than 50% in each case. So it's certainly not full but it's
a long way from empty.

One bit of good news: I've got quite a lot of room on both of the 750 GB
drives if we can figure out how to copy the data from the 3 TB to the
smaller drives. Out of the usable 698 GB on both smaller drives, I have
549 GB free on the C: and 562 GB free on the D:. Any idea how we can
copy the data off the 3 TB to the smaller drives?

If I could see the full folders, I could pick and choose the data that I
have on the 3 TB; some of it is definitely less crucial than others and
I could lose the less crucial stuff with a lot less pain than some of
the others.

I should also mention that I have acquired a basic familiarity with
Linux/Unix over the years so I'm not particularly intimidated by using
that OS if I need to. I'm nowhere near an expert and I'm pretty rusty
even about the things I do know but I'm game to try things.
--
Rhino

Paul

unread,

May 23, 2013, 8:32:38 PM5/23/13

to

Rhino wrote:

<<slight snip>>

> I hadn't looked at the disk-available/disk-used numbers in a few months
> and don't remember them clearly in any case. I know the three partitions
> were definitely NOT full but each had a fair percentage of data on it,
> probably more than 50% in each case. So it's certainly not full but it's
> a long way from empty.
>
> One bit of good news: I've got quite a lot of room on both of the 750 GB
> drives if we can figure out how to copy the data from the 3 TB to the
> smaller drives. Out of the usable 698 GB on both smaller drives, I have
> 549 GB free on the C: and 562 GB free on the D:. Any idea how we can
> copy the data off the 3 TB to the smaller drives?
>
> If I could see the full folders, I could pick and choose the data that I
> have on the 3 TB; some of it is definitely less crucial than others and
> I could lose the less crucial stuff with a lot less pain than some of
> the others.
>
> I should also mention that I have acquired a basic familiarity with
> Linux/Unix over the years so I'm not particularly intimidated by using
> that OS if I need to. I'm nowhere near an expert and I'm pretty rusty
> even about the things I do know but I'm game to try things.

I don't even know what an EPBR is :-)

====================================================================
Drive 3 (764428 MB) - 97451 cyl, 255 heads, 63 sectors per track

Partition Table at sector 0 (cyl 0, head 0, sector 1):

STARTING ENDING SECTORS
TYPE BOOT CYL HEAD SECTOR CYL HEAD SECTOR BEFORE SECTORS

1 05 80 1 0 1 1023 254 63 16065 1565534250 <-- 746.5GB
2 07 00 1023 254 63 1023 254 63 156555031 2729411370 <-- 1301.5GB

3 00 00 0 0 0 0 0 0 0 0
4 00 00 0 0 0 0 0 0 0 0
======================================================================

When I click the "Goto EPBR" button, the display changes to this:

====================================================================
Drive 3 (764428 MB) - 97451 cyl, 255 heads, 63 sectors per track

Partition Table at sector 16065 (cyl 1, head 0, sector 1):

STARTING ENDING SECTORS
TYPE BOOT CYL HEAD SECTOR CYL HEAD SECTOR BEFORE SECTORS

1 07 00 1 1 1 1023 254 63 63 1565534187 <-- 746.5GB

2 00 00 0 0 0 0 0 0 0 0
3 00 00 0 0 0 0 0 0 0 0
4 00 00 0 0 0 0 0 0 0 0
======================================================================

*******

Oh, dear. A new curve ball. What exactly is TYPE=05 ???

http://www.win.tue.nl/~aeb/partitions/partition_types-1.html

05 DOS 3.3+ Extended Partition

Really ? Where the hell did that come from ?
What tool did that ?

Paul

Paul

unread,

May 23, 2013, 9:34:51 PM5/23/13

to

OK, took a blank 250GB disk, and created some partitions.

http://imageshack.us/a/img515/3900/logicalntfsinextended.gif

I first created an Extended partition. That has type 0x0F.
(Whereas yours is 0x05, but much the same thing.)

Inside it, I created an NTFS partition of around 20GB. It
has type 0x07. Clicking the EPBR button, and that's when it
told me what logical partition(s) were inside the Extended.

So it's "normal", but not something you want to do if you
want easy data management. If I create an extended one
on a disk, I like to make it down on the right hand end.
I don't normally use them, because they make data
recovery harder (grrr...).

Anyway, because my mind drifts easily, I should get back to
the point.

Your results show me what is happening to the lower 2.2TB
part. Your upper portion, would be under a different
drive number in PTEDIT32. For example, when I used the
Acronis driver here, Disk 3 was the 2.2TB portion,
while Disk 5 was the upper portion (total being 3TB).
So with the Acronis driver, it turns one drive, into
two drives. And those drives happened to be numbered
Disk 3 and Disk 5 here. Your "Disk 5" equivalent, you
haven't reported it yet.

If the Acronis driver is not working, or some of the
related services, maybe that's why the upper portion
is missing. Remember my first results, where an old,
stale Acronis driver, prevented me from seeing the
upper portion. I only got to see Drive 5, when a
clean (old) OS was loaded, and a new Acronis driver
was installed.

Now, given this new information, not much has changed.

+------------------------------------+------------------------------------+
| Extended 0x05 type, 1565534250 sect| Primary 0x07 NTFS, 2729411370 sect |
+--------------------------------+---+------------------------------------+
| Logical 0x07 NTFS, 1565534187 |
+--------------------------------+

So this error is coming from the beginning of the right-most partition,
not near its end. Curious. What does that mean ?

"Error: Failed to read from sector '1,571,841,771' of hard disc '3'"

1571841771-1565534250 = 6307521, 6307521 * 512 = 3,229,450,752 bytes.

About 3GB from the beginning.

Well, I've got no theory for this... Hmmm.

I'd try a bad block scan from HDTune. Probably take around 5 hours or so.
It should report it's attempting to scan the visible 2.2TB of stuff.

There's no reason except a busted sector, for it to be dying at the
3GB mark, into the Primary partition.

Now, at this point, TestDisk would also be an option. (Don't update the
partition table, just see if you can access files on the primary partition.)
I think TestDisk can do that, without the partition being mounted. And
I suspect it is failing to mount right now.

http://www.cgsecurity.org/wiki/TestDisk_Step_By_Step

Of course, if TestDisk actually runs into bad sectors, it could
come to a grinding halt all on its own. TestDisk skips along
at regular intervals - it doesn't read the entire disk, it goes
like a "picket fence", looking for partitions. If there is actually
a bad spot in the disk, this could be much worse than I originally
suspected.

See, with TestDisk, you can actually look at file structure,
assuming things aren't busted enough to stop it.

http://www.cgsecurity.org/mw/images/List_files.gif

If I haven't blown it again, this is looking like a
serious data recovery problem. No longer "fun and games".

Paul

Paul

unread,

May 24, 2013, 1:38:51 AM5/24/13

to

>> Rhino wrote:

>>> I hadn't looked at the disk-available/disk-used numbers in a few
>>> months and don't remember them clearly in any case. I know the three
>>> partitions were definitely NOT full but each had a fair percentage of
>>> data on it, probably more than 50% in each case. So it's certainly
>>> not full but it's a long way from empty.
>>>
>>> One bit of good news: I've got quite a lot of room on both of the 750
>>> GB drives if we can figure out how to copy the data from the 3 TB to
>>> the smaller drives. Out of the usable 698 GB on both smaller drives,
>>> I have 549 GB free on the C: and 562 GB free on the D:. Any idea how
>>> we can copy the data off the 3 TB to the smaller drives?
>>>
>>> If I could see the full folders, I could pick and choose the data
>>> that I have on the 3 TB; some of it is definitely less crucial than
>>> others and I could lose the less crucial stuff with a lot less pain
>>> than some of the others.
>>>
>>> I should also mention that I have acquired a basic familiarity with
>>> Linux/Unix over the years so I'm not particularly intimidated by
>>> using that OS if I need to. I'm nowhere near an expert and I'm pretty
>>> rusty even about the things I do know but I'm game to try things.

One other note. I just tested a 3TB hard drive, on a USB
adapter, and I got a report that the disk was 800GB long.
Windows would not mount the partition, and it was listed
as "raw" when I did properties on it. So the adapter
seemed to be reporting the wrong disk size.

It returned back to normal (one large 2.2TB NTFS partition),
once it was plugged back into a SATA port on my current OS.
So no damage done.

This is not to say that all adapters do this. I can only
state it did that for my adapter. Didn't matter whether
Acronis driver was present or not. Even Linux reported
it the same way.

*******

An adapter here is listed as currently supporting 2TB.

http://www.newegg.com/Product/Product.aspx?Item=N82E16812232002

Talk about weird. This one is rated "up to 3TB". Why didn't
they test it with a 4TB drive ? The 4TB drives are in the market.

http://www.newegg.com/Product/Product.aspx?Item=N82E16812232033

Paul

Rhino

unread,

May 24, 2013, 5:55:15 PM5/24/13

to

On 2013-05-23 20:32, Paul wrote:
> Rhino wrote:
>
> <<slight snip>>
>
>> I hadn't looked at the disk-available/disk-used numbers in a few
>> months and don't remember them clearly in any case. I know the three
>> partitions were definitely NOT full but each had a fair percentage of
>> data on it, probably more than 50% in each case. So it's certainly not
>> full but it's a long way from empty.
>>
>> One bit of good news: I've got quite a lot of room on both of the 750
>> GB drives if we can figure out how to copy the data from the 3 TB to
>> the smaller drives. Out of the usable 698 GB on both smaller drives, I
>> have 549 GB free on the C: and 562 GB free on the D:. Any idea how we
>> can copy the data off the 3 TB to the smaller drives?
>>
>> If I could see the full folders, I could pick and choose the data that
>> I have on the 3 TB; some of it is definitely less crucial than others
>> and I could lose the less crucial stuff with a lot less pain than some
>> of the others.
>>
>> I should also mention that I have acquired a basic familiarity with
>> Linux/Unix over the years so I'm not particularly intimidated by using
>> that OS if I need to. I'm nowhere near an expert and I'm pretty rusty
>> even about the things I do know but I'm game to try things.
>
> I don't even know what an EPBR is :-)
>

Me neither! I swear I haven't googled on EPBR but here's a guess that
actually seems plausible: Extended Partition Boot Record. It just came
to me as I tried to compose a wildly improbable meaning just for a
laugh....

PTEDIT32. And no, I don't know what version of the program it is. It
doesn't say in either the program itself or the file I downloaded. The
page I downloaded it from -
http://www.pendriveapps.com/partition-table-editor-ptedit/ - doesn't
state a version number either. It may be a very old version. I had a lot
of trouble finding that program. I was surprised to NOT find it at the
Symantec site. If it was one of theirs, you'd think they'd give you a
way to download it.

--
Rhino

Rhino

unread,

May 24, 2013, 6:36:25 PM5/24/13

to

On 2013-05-23 21:34, Paul wrote:
> Paul wrote:
>> Rhino wrote:
>>
>> <<slight snip>>
>>
>>> I hadn't looked at the disk-available/disk-used numbers in a few
>>> months and don't remember them clearly in any case. I know the three
>>> partitions were definitely NOT full but each had a fair percentage of
>>> data on it, probably more than 50% in each case. So it's certainly
>>> not full but it's a long way from empty.
>>>
>>> One bit of good news: I've got quite a lot of room on both of the 750
>>> GB drives if we can figure out how to copy the data from the 3 TB to
>>> the smaller drives. Out of the usable 698 GB on both smaller drives,
>>> I have 549 GB free on the C: and 562 GB free on the D:. Any idea how
>>> we can copy the data off the 3 TB to the smaller drives?
>>>
>>> If I could see the full folders, I could pick and choose the data
>>> that I have on the 3 TB; some of it is definitely less crucial than
>>> others and I could lose the less crucial stuff with a lot less pain
>>> than some of the others.
>>>
>>> I should also mention that I have acquired a basic familiarity with
>>> Linux/Unix over the years so I'm not particularly intimidated by
>>> using that OS if I need to. I'm nowhere near an expert and I'm pretty
>>> rusty even about the things I do know but I'm game to try things.
>>
>> I don't even know what an EPBR is :-)
>>

As I mentioned in the other thread, I'm guessing Extended Partition Boot
Record. But that really IS a guess!

Oh, what the heck, let's google this and see.... How about that? A
simple Google of EPBR results in a hit at
http://www.acronymfinder.com/EPBR.html
which gives two possible meanings, one of which is indeed Extended
Partition Boot Record. And this site
http://www.applexsoft.com/glossary/epbr.html
actually DEFINES it.

PTEDIT32. I have no idea which version though.

>
> OK, took a blank 250GB disk, and created some partitions.
>
> http://imageshack.us/a/img515/3900/logicalntfsinextended.gif
>
> I first created an Extended partition. That has type 0x0F.
> (Whereas yours is 0x05, but much the same thing.)
>
> Inside it, I created an NTFS partition of around 20GB. It
> has type 0x07. Clicking the EPBR button, and that's when it
> told me what logical partition(s) were inside the Extended.
>
> So it's "normal", but not something you want to do if you
> want easy data management. If I create an extended one
> on a disk, I like to make it down on the right hand end.
> I don't normally use them, because they make data
> recovery harder (grrr...).
>

I don't know if I consciously chose to create an extended partition or
not when I set up the 3 TB drive. I've never really understood most of
the concepts around organizing disks. Maybe I made some bad decisions;
maybe the program made that decision for me. I simply don't recall. In
any case, it's water under the bridge at this point....

> Anyway, because my mind drifts easily, I should get back to
> the point.
>
> Your results show me what is happening to the lower 2.2TB
> part. Your upper portion, would be under a different
> drive number in PTEDIT32. For example, when I used the
> Acronis driver here, Disk 3 was the 2.2TB portion,
> while Disk 5 was the upper portion (total being 3TB).
> So with the Acronis driver, it turns one drive, into
> two drives. And those drives happened to be numbered
> Disk 3 and Disk 5 here. Your "Disk 5" equivalent, you
> haven't reported it yet.

PTEDIT32 shows me exactly 3 disks in the drop-down that lists the disks.
The first two of them are the exact same size (715402 MB). I would just
about bet my life that those are the two smaller drives, which are
nominally 750 GB drives, although you never get all of that to play
with. The data I reported to you was from the remaining drive, which is
the 3 TB one.

>
> If the Acronis driver is not working, or some of the
> related services, maybe that's why the upper portion
> is missing. Remember my first results, where an old,
> stale Acronis driver, prevented me from seeing the
> upper portion. I only got to see Drive 5, when a
> clean (old) OS was loaded, and a new Acronis driver
> was installed.
>
> Now, given this new information, not much has changed.
>
> +------------------------------------+------------------------------------+
> | Extended 0x05 type, 1565534250 sect| Primary 0x07 NTFS, 2729411370 sect |
> +--------------------------------+---+------------------------------------+
> | Logical 0x07 NTFS, 1565534187 |
> +--------------------------------+
>
> So this error is coming from the beginning of the right-most partition,
> not near its end. Curious. What does that mean ?
>
> "Error: Failed to read from sector '1,571,841,771' of hard disc '3'"
>
> 1571841771-1565534250 = 6307521, 6307521 * 512 = 3,229,450,752 bytes.
>
> About 3GB from the beginning.
>
> Well, I've got no theory for this... Hmmm.
>
> I'd try a bad block scan from HDTune. Probably take around 5 hours or so.
> It should report it's attempting to scan the visible 2.2TB of stuff.
>

I got HDTune a while back when someone on the newsgroup suggested it -
maybe you! I just launched it and it is showing my 3 TB drive as having
801 GB. I selected that drive and then went to the Error Scan tab. Since
you're saying it will run 5 hours or more, I assume that is NOT the
quick scan so I left that unchecked and clicked start. I'll report back
when I see what it's done....

> There's no reason except a busted sector, for it to be dying at the
> 3GB mark, into the Primary partition.
>
> Now, at this point, TestDisk would also be an option. (Don't update the
> partition table, just see if you can access files on the primary
> partition.)
> I think TestDisk can do that, without the partition being mounted. And
> I suspect it is failing to mount right now.
>
> http://www.cgsecurity.org/wiki/TestDisk_Step_By_Step
>

I'm downloading it (Version 6.13) as I run this. I assume it would be
bad to run it will running HDTune (and downloading files on the other
two drives) so I'll wait until all that stuff is done before attempting it.

> Of course, if TestDisk actually runs into bad sectors, it could
> come to a grinding halt all on its own. TestDisk skips along
> at regular intervals - it doesn't read the entire disk, it goes
> like a "picket fence", looking for partitions. If there is actually
> a bad spot in the disk, this could be much worse than I originally
> suspected.
>
> See, with TestDisk, you can actually look at file structure,
> assuming things aren't busted enough to stop it.
>
> http://www.cgsecurity.org/mw/images/List_files.gif
>
> If I haven't blown it again, this is looking like a
> serious data recovery problem. No longer "fun and games".
>

You mean it's been fun this whole time and is only just starting to turn
ugly? Jeez, I thought it was pretty hellish from the start ;-)

--
Rhino

Rhino

unread,

May 24, 2013, 6:44:35 PM5/24/13

to

Could my problem REALLY be resolved by simply powering the computer off
(ALL the way off this time!) and swapping the SATA connectors around?
Because if that holds any prospect of fixing everything easily, I'm all
for it!! Seriously!

That seems very safe and quick so I would gladly do it if it actually
has some chance of working. I could just label each SATA cable with its
role, then swap them from drive to drive then boot up and see if my E:
F: and G: drives are back. It would take a few tries to hit all the
combinations but I wouldn't mind doing that. I'd document what I was
doing as well so that I'm sure I'm doing different combinations each
time. I really don't remember what role (Master, Slave, Primary,
Secondary) I assigned to each device, unfortunately, otherwise it would
be even quicker. But I'll try every combination and be done MUCH faster
than having to guess my way through the other things we've been talking
about.

--
Rhino

Paul

unread,

May 24, 2013, 7:54:06 PM5/24/13

to

Rhino wrote:

>
> Could my problem REALLY be resolved by simply powering the computer off
> (ALL the way off this time!) and swapping the SATA connectors around?
> Because if that holds any prospect of fixing everything easily, I'm all
> for it!! Seriously!
>
> That seems very safe and quick so I would gladly do it if it actually
> has some chance of working. I could just label each SATA cable with its
> role, then swap them from drive to drive then boot up and see if my E:
> F: and G: drives are back. It would take a few tries to hit all the
> combinations but I wouldn't mind doing that. I'd document what I was
> doing as well so that I'm sure I'm doing different combinations each
> time. I really don't remember what role (Master, Slave, Primary,
> Secondary) I assigned to each device, unfortunately, otherwise it would
> be even quicker. But I'll try every combination and be done MUCH faster
> than having to guess my way through the other things we've been talking
> about.
>

Could you do me a favor ?

Could you fire up PTEDIT32 again, and
list *all* the disk entries. Using that
info, I may be able to satisfy myself as
to what is appearing and what is not appearing.
For example, if I see a 2TB entry, I would be
expecting to see the 746GB entry further down
the list of disks. And the disks listed in PTEDIT32,
should be in the same order as you see them in Disk Management.

Disk 1 MBR
Disk 2 MBR
EPBR or whatever
Disk 3 MBR
...

As I've seen the "only 800GB showing up" thing here,
you can't really expect TestDisk to perform miracles,
when Windows is preventing it from examining the whole 3TB disk.
Again, this is related to the Acronis TI software
somehow.

And I explained, in my "experiment", how, worst case,
I'm able to examine above the 2.2TB mark. By firing up
Linux, I can't "go above" 2.2TB with regular file system
operations. But, if I use "dd", it can read the sectors
above 2.2TB. I was able to copy the 746GB of sectors above 2.2TB,
then bring them down below 2.2TB, mount them, and get the
files off. That's currently the only way I know of, to
do data recovery on the disk, with any assurance it will
work. If I was counting on Acronis, I'd be counting on
a pretty weak solution.

My batting average on seeing the entire 3TB disk, is
pretty bad now. I would not be surprised if yours
was in the same mess. For me to see the whole disk
now, I have to go back to my 2011 backup copy of WinXP
again.

The fact you used Extended/Logical, is not the end
of the world. It just complicates recovery later.

Once I see all your disk entries, at least then I'll know
whether Acronis is working right now or not. As of right
now, I suspect there is an issue with the hard drive,
at a point about 3GB from the left hand end of the
partition.

Disk "N"
Partition 1 Partition 2
^
Disk "N+1" |
Partition 3 Broken, here...

Paul

Rhino

unread,

May 24, 2013, 8:56:32 PM5/24/13

to

On 2013-05-24 19:54, Paul wrote:
> Rhino wrote:
>
>>
>> Could my problem REALLY be resolved by simply powering the computer
>> off (ALL the way off this time!) and swapping the SATA connectors
>> around? Because if that holds any prospect of fixing everything
>> easily, I'm all for it!! Seriously!
>>
>> That seems very safe and quick so I would gladly do it if it actually
>> has some chance of working. I could just label each SATA cable with
>> its role, then swap them from drive to drive then boot up and see if
>> my E: F: and G: drives are back. It would take a few tries to hit all
>> the combinations but I wouldn't mind doing that. I'd document what I
>> was doing as well so that I'm sure I'm doing different combinations
>> each time. I really don't remember what role (Master, Slave, Primary,
>> Secondary) I assigned to each device, unfortunately, otherwise it
>> would be even quicker. But I'll try every combination and be done MUCH
>> faster than having to guess my way through the other things we've been
>> talking about.
>>
>
> Could you do me a favor ?

Sure. But just before I do that, HDTune is already done. It took just
69:58 minutes. Of course, I think it only read 800 GB of the 3 TB drive.
However, the Error Scan found no bad blocks: every block in the graphic
was green. The Position value at the end was 764130 MB.

>
> Could you fire up PTEDIT32 again, and
> list *all* the disk entries.

I tried launching PTEDIT32 again but it gives me an error dialog:
"Error - Error #5 starting the PowerQuest Engine."

I thought it was because I was running a lengthy batch process that
affects the D: drive so I paused that and tried again. It still doesn't
work though. The display of the disk information appears for a split
second, then disappears again. I think I'm going to have to reboot and
see what happens then. I'm going to finish my batch process first, then
reboot and do the PTEDIT again and report back. That'll be a few hours
I expect. ...

--
Rhino

Paul

unread,

May 24, 2013, 9:08:43 PM5/24/13

to

Rhino wrote:
> On 2013-05-24 19:54, Paul wrote:

>> Could you fire up PTEDIT32 again, and
>> list *all* the disk entries.
>
> I tried launching PTEDIT32 again but it gives me an error dialog:
> "Error - Error #5 starting the PowerQuest Engine."
>
> I thought it was because I was running a lengthy batch process that
> affects the D: drive so I paused that and tried again. It still doesn't
> work though. The display of the disk information appears for a split
> second, then disappears again. I think I'm going to have to reboot and
> see what happens then. I'm going to finish my batch process first, then
> reboot and do the PTEDIT again and report back. That'll be a few hours
> I expect. ...
>

In Windows 7, you need to find a way to "Run As Administrator".
You can type "cmd" in Start, select Run As Administrator by right-clicking
on it, then that Command Prompt window is elevated. And programs
run from there, run as administrator. Similarly, if you can figure out
how, you can do Run As Administrator, directly on the PTEDIT32 program.

When that happened to me, I saw the "Error 5", but it didn't take
too long to figure out it was a permissions problem. You can't
do things like access an MBR, without being an Administrator.
Same thing happens on Linux, when you run "dd", "fdisk", and the
like. They all require administrator (root) privilege.

Paul

Rhino

unread,

May 25, 2013, 10:06:14 AM5/25/13

to

On 2013-05-24 21:08, Paul wrote:

> Rhino wrote:
>> On 2013-05-24 19:54, Paul wrote:
>
>>> Could you fire up PTEDIT32 again, and
>>> list *all* the disk entries.
>>
>> I tried launching PTEDIT32 again but it gives me an error dialog:
>> "Error - Error #5 starting the PowerQuest Engine."
>>
>> I thought it was because I was running a lengthy batch process that
>> affects the D: drive so I paused that and tried again. It still
>> doesn't work though. The display of the disk information appears for a
>> split second, then disappears again. I think I'm going to have to
>> reboot and see what happens then. I'm going to finish my batch process
>> first, then reboot and do the PTEDIT again and report back. That'll be
>> a few hours
>> I expect. ...
>>
>
> In Windows 7, you need to find a way to "Run As Administrator".
> You can type "cmd" in Start, select Run As Administrator by right-clicking
> on it, then that Command Prompt window is elevated. And programs
> run from there, run as administrator. Similarly, if you can figure out
> how, you can do Run As Administrator, directly on the PTEDIT32 program.
>

I only have have two user accounts on the XP machine and both are
already administrators. I had been simply clicking on the zip file, then
clicking on PTEDIT32.EXE, which is alone in the zip file, but that
started producing the "Error 5" dialogs. After ruling out the
administrator issue, I simply extracted PTEDIT32.EXE from the zip file
and put it on my C: drive and created a shortcut for the desktop. Now,
when I click on the shortcut, it works just fine. I'll report the result
back where you'd asked me for the favour.

> When that happened to me, I saw the "Error 5", but it didn't take
> too long to figure out it was a permissions problem. You can't
> do things like access an MBR, without being an Administrator.
> Same thing happens on Linux, when you run "dd", "fdisk", and the
> like. They all require administrator (root) privilege.
>
> Paul

--
Rhino

Rhino

unread,

May 25, 2013, 10:17:26 AM5/25/13

to

On 2013-05-24 19:54, Paul wrote:

> Rhino wrote:
>
>>
>> Could my problem REALLY be resolved by simply powering the computer
>> off (ALL the way off this time!) and swapping the SATA connectors
>> around? Because if that holds any prospect of fixing everything
>> easily, I'm all for it!! Seriously!
>>
>> That seems very safe and quick so I would gladly do it if it actually
>> has some chance of working. I could just label each SATA cable with
>> its role, then swap them from drive to drive then boot up and see if
>> my E: F: and G: drives are back. It would take a few tries to hit all
>> the combinations but I wouldn't mind doing that. I'd document what I
>> was doing as well so that I'm sure I'm doing different combinations
>> each time. I really don't remember what role (Master, Slave, Primary,
>> Secondary) I assigned to each device, unfortunately, otherwise it
>> would be even quicker. But I'll try every combination and be done MUCH
>> faster than having to guess my way through the other things we've been
>> talking about.
>>
>
> Could you do me a favor ?
>
> Could you fire up PTEDIT32 again, and
> list *all* the disk entries.

As requested:

====================================================================
Drive 1 (715402 MB) - 91201 cyl, 255 heads, 63 sectors per track

Partitition Table at sector 0 (cyl 0, head 0, sector 1):

STARTING ENDING SECTORS
TYPE BOOT CYL HEAD SECTOR CYL HEAD SECTOR BEFORE SECTORS

1 07 00 0 1 1 1023 254 63 63 1465144002

2 00 00 0 0 0 0 0 0 0 0
3 00 00 0 0 0 0 0 0 0 0
4 00 00 0 0 0 0 0 0 0 0
======================================================================

====================================================================
Drive 2 (715402 MB) - 91201 cyl, 255 heads, 63 sectors per track

Partitition Table at sector 0 (cyl 0, head 0, sector 1):

STARTING ENDING SECTORS
TYPE BOOT CYL HEAD SECTOR CYL HEAD SECTOR BEFORE SECTORS

1 07 80 0 1 1 1023 254 63 63 1465127937

2 00 00 0 0 0 0 0 0 0 0
3 00 00 0 0 0 0 0 0 0 0
4 00 00 0 0 0 0 0 0 0 0
======================================================================

The only differences I can see between Drive 1 and 2 are that Drive 2's
entry has a Boot value of 80 instead of 0 and the number of sectors is
slightly smaller on Drive 2 (some bad sectors somewhere?).

Here's the stuff from Drive 3 again, just so that you have it all handy
on the same page:

====================================================================
Drive 3 (764428 MB) - 97451 cyl, 255 heads, 63 sectors per track

Partitition Table at sector 0 (cyl 0, head 0, sector 1):

STARTING ENDING SECTORS
TYPE BOOT CYL HEAD SECTOR CYL HEAD SECTOR BEFORE SECTORS
1 05 80 1 0 1 1023 254 63 16065 1565534250

2 07 00 1023 254 63 1023 254 63 156555031 2729411370

3 00 00 0 0 0 0 0 0 0 0
4 00 00 0 0 0 0 0 0 0 0
======================================================================

When I click the "Goto EPBR" button, the display changes to this:

====================================================================
Drive 3 (764428 MB) - 97451 cyl, 255 heads, 63 sectors per track

Partitition Table at sector 16065 (cyl 1, head 0, sector 1):

STARTING ENDING SECTORS
TYPE BOOT CYL HEAD SECTOR CYL HEAD SECTOR BEFORE SECTORS
1 07 00 1 1 1 1023 254 63 63 1565534187

2 00 00 0 0 0 0 0 0 0 0
3 00 00 0 0 0 0 0 0 0 0
4 00 00 0 0 0 0 0 0 0 0
======================================================================

--
Rhino

Paul

unread,

May 25, 2013, 10:42:21 AM5/25/13

to

One disk, the partition is one cylinder shorter than the other disk.
(255*63 sectors). I'm assuming Disk 1 and Disk 2 in this case,
are two separate, non-3TB drives.

On your 3TB (Disk 3), I would have expected the Acronis driver, to make
the Disk 3 like this, plus the "fake" Disk 4 containing the rest
of the space above the 2.2TB mark. The Acronis driver, converts
one physical disk, into two disks as far as Disk Management
and other tools are concerned.

Disk 3

+------------------------------------+------------------------------------+
| Extended 0x05 type, 1565534250 sect| Primary 0x07 NTFS, 2729411370 sect |
+--------------------------------+---+------------------------------------+
| Logical 0x07 NTFS, 1565534187 |
+--------------------------------+

Disk 4

+------------------------------------+
| |
+------------------------------------+

So an Acronis driver malfunction, makes Disk 4 disappear. And
you report a read error, while accessing the right-most partition
on Disk 3. To get Disk 4 back, you'd need some way to fix the
Acronis problem. I doubt reinstalling Acronis would help, and you've
probably already tried that. In my case, using a cleaner
copy of WinXP from two years ago, then installing Acronis,
I got the upper section to appear.

Note that, when Acronis prepares the disk (the Extended Capacity Manager
allocate function), it creates a 262144 byte section between Disk 3 and
Disk 4, and that should contain the information Acronis might use at
the driver level. The layout of Disk 4 is conventional, and I was
able to copy the Disk 4 type area, from Linux, to do data recovery
on it.

Paul

Rhino

unread,

May 25, 2013, 11:08:00 AM5/25/13

to

No, I haven't tried removing Acronis yet. I'd still be willing to try it
but, as I mentioned earlier, I am doubtful that I will remove all the
"bits and pieces" correctly since the description of what to do and how
to do it is rather vague in places. Removing the program itself seems
straightforward and simply uninstalling the two Acronis drivers from
Device Manager sounds simple on paper (although I doubt it will be given
all the grief it has given me about the sound drivers). I'm a lot more
worried about removing all the registry entries and services correctly
(and in the correct order). It appalls me that Acronis itself seems to
be useless at guiding people through this process. If they can't even
figure out how to remove their own, er, stuff, how on earth are guys
like me supposed to figure it out?

But if you want me to try, I would at least consider it.

> In my case, using a cleaner
> copy of WinXP from two years ago, then installing Acronis,
> I got the upper section to appear.
>

I've still got my original XP CD and I have the image I created of XP
SP2 which I installed to make the jump from XP to XP SP2. In doing the
upgrade to SP3, I did not take a full backup, although I did back up
SOME things (email and such). I did the jump to SP3 after buying the 3
TB drive but before I'd physically installed the drive. I don't remember
how I did the upgrade to SP3 via Windows Updates and don't have an image
of the computer at that time. Does any of that help us?

> Note that, when Acronis prepares the disk (the Extended Capacity Manager
> allocate function), it creates a 262144 byte section between Disk 3 and
> Disk 4, and that should contain the information Acronis might use at
> the driver level. The layout of Disk 4 is conventional, and I was
> able to copy the Disk 4 type area, from Linux, to do data recovery
> on it.
>

I was just about to start running TestDisk but I'm having trouble
finding the post where you described what to do with it. I've been
running in circles for the last hour trying to figure out how to track
down all the posts in this thread containing "TestDisk" in the body but
the Thunderbird filters and search facility are so cryptic that I'm
having trouble finding it. (I thought I'd spend a few minutes learning
how to use that feature of Thunderbird and consider it an investment in
the future; instead, I'm not finding anything and have to go back to
reading each (recent) post in this thread one at a time to find the bit
that I want, which wasn't very detailed anyway as I recall.)

Can you recap just what I'm going to do with TestDisk once I've selected
the 3 TB drive? (By the way, TestDisk also describes the 3 TB drive as
being 801 GB).

Also, I still have that Unbuntu disk here on my desk and am ready to use
it. I could easily imagine spending a big part of today grabbing chunks
of the data on my 3 TB drive and copying it off to the smaller drives
and/or burning it to disk. (I'm also tempted to see if it would reveal
anything about the sound problem but I'm afraid to get sidetracked on
that since it is more of a serious annoyance while I REALLY want to get
that data safe and secure.)

Will it be self-evident how to use the Ubuntu CD once I reboot with it
in the DVD drive so that it only inspects my system WITHOUT installing
itself and wiping out Windows? I'm hoping it has a menu that says (in
effect): "Run Ubuntu from this CD without touching your native OS" -OR-
"Install Ubuntu on your computer overtop of your existing OS". I'm a
little worried that if you don't start the disk in some non-intuitive
way, like from a command prompt with a specific command, it will just
install Ubuntu over Windows and then really put me in the soup....

--
Rhino

Paul

unread,

May 25, 2013, 12:02:26 PM5/25/13

to

There is almost no point in running TestDisk at this point.

The only point in running TestDisk (from Windows), is if Disk 3 is
being reported as a 2.2TB disk. You recently ran HDTune, and the bad
block scan only spanned 800GB. Which means, Windows is preventing any
operation above the 800GB point. If you cannot get above the 800GB
point, "no way or no how", then TestDisk will run into the same limit.
If it can't see the whole disk, it's useless.

TestDisk can also be run from linux, from a Linux LiveCD. You
may even find a copy of TestDisk is standard on the LiveCD (doesn't
require a trip to Synaptic Package Manager, to download a copy).
So you can try that if you want.

If Linux has problems reading the upper NTFS partition, it will
likely refuse to mount it as well.

So the only difference at the moment, is Windows isn't going to
do anything good, until driver issues are resolved. While on
Linux, you won't be able to see the upper portion (>2.2TB) of the
3TB disk in any case (unless Acronis makes a Linux package that is).
And if you run TestDisk from Linux, you might be able to see the
right-most partition of Disk 3 and some of the files on it.

http://www.cgsecurity.org/wiki/TestDisk_Step_By_Step

In this example photo, you can see TestDisk being used to list
files on a discovered partition. That's all I really wanted
to see from TestDisk, that you can still see files.

http://www.cgsecurity.org/mw/images/List_files.gif

I wouldn't want you to actual make changes from TestDisk,
because at this point, there is no evidence of MBR damage.
I think your partition information is correct. You may
have a physical problem with the disk, about 3GB into the
right-most partition. And nothing is going to fix that,
short of more data recovery work (and a second disk drive
to hold the results). Since your partitions are occupied
with files, there probably isn't room to attempt
data recovery with the existing disk (the right-most partition
is larger than the others).

When in HDTune, have you checked the Health tab ?
What does it show for Disk 3 ? Any "yellow marks" in there,
like for "Reallocated Sectors" or "Current Pending" ? A non-zero
data field there, would be a lead indicator the drive isn't
healthy.

Paul

Rhino

unread,

May 25, 2013, 12:05:35 PM5/25/13

to

On 2013-05-25 10:42, Paul wrote:

When I fired up TestDisk, I told it to create a new log file on the
first page. When I got to the second page, it listed the disk drives and
shows the big drive as "Disk /dev/sdc - 801 GB / 746 GiB - ST3000DM
001-9YN166". The note on the bottom of the page is VERY worrisome: "Disk
capacity much be correctly detected for a successful recovery. If a disk
listed above has incorrect size, check HD jumper settings, BIOS
detection, and install the latest OS patches and disk drivers."

Now, if I'm only using this program to LOOK AT the big drive then it may
be safe to proceed but I have no idea if proceeding past this point will
actually cause some kind of write to the hard drive that mangles it
irretrievably.

As for the suggestions in the note, it seems very unlikely that the tech
changed the jumpers on the hard drive (I don't remember if it even HAD
any jumpers). However, it's possible that the BIOS simply isn't set
correctly to recognize the large drive. Any idea what I'm looking for in
the ASrock BIOS that would make more of the large drive visible? I was
in there the other day and nothing jumped out at me but I find the BIOS
a fairly cryptic place much of the time anyway. I could easily have
missed something.

[Crosses fingers] Okay, I went ahead and clicked Proceed, then chose
Intel as the partition type, the chose Analyse. The following is displayed:
======================================================================
Disk /dev/sdc - 810 GB / 746 GiB - CHS 97451 255 63
Current partition structure:
Partition Start END Size in Sectors
1 E extended 1 0 1 97450 254 63 1565534250
2 P HPFS - NTSF 97451 0 1 267348 254 63 2729411370
No partition is bootable
5 L HPFS - NTFS 1 1 1 97450 254 63 1565534187 [Foo]

*=Primary bootable P=Primary L=Logical E=Extended D=Deleted
>[QUick Search] [ Backup ]
Try to locate partition
=======================================================================

Foo in the report above is just a placeholder for the Windows name of
the F: drive, one of the three Windows drives that were visible when
this drive was fully visible to the computer on the old motherboard.

There's no button or option to go back or up in the menu hierarchy so I
simply closed the Command window to assure that I didn't accidentally
caused a write.

Does any of this help? If so, what should I do next? I'm tempted to
reboot with the Ubuntu CD to see if will show us all or most of the data
on the 3 TB drive. Then I can start copying it to my smaller hard
drives, DVD-RWs, or even to my laptop (which has lots of space on its
hard drives) if you can suggest a way to get it from one computer to the
other.

I could upload it from the desktop to a server containing free hosting
and then download it to the laptop but I think the limits on space on
the free hosting account are pretty severe. I'd probably have to
transfer the data a couple of GB at a time. But first, help me get
access to the data. Then we'll figure out how to get it to someplace safer.

--
Rhino

Paul

unread,

May 25, 2013, 12:50:32 PM5/25/13

to

If the drive capacity is listed as 800GB, and TestDisk has even given
you a warning, there's no point going much further with TestDisk under
Windows.

It's of minor interest, to use TestDisk on a platform that reports
Disk 3 as being 2.2TB, as then, you can view files. I.e. The partition
still doesn't mount (because it is broken), but TestDisk can attempt
to use what information is there, to display some of the files. Copying
files from the damaged section, might still be impossible in any case.

Your priorities:

1) Come up with an environment, where the Acronis driver can turn
Disk 3, into Disk 3 and Disk 4, and give you back access to the data.

2) There is a minor possibility the status of the right-most partition
(broken), might change during step 1. But if not, you *still* have to
complete step 1, in order for *any* utility to recover the files
off the Disk 3 right-most partition. As long as that disk is stuck
at reporting 800GB, you're not going to get any access to the files.

HTH,
Paul

Rhino

unread,

May 25, 2013, 1:27:31 PM5/25/13

to

On 2013-05-25 12:02, Paul wrote:
> Rhino wrote:
>> On 2013-05-25 10:42, Paul wrote:
>>> Rhino wrote:
>>>> On 2013-05-24 19:54, Paul wrote:
>>>>> Rhino wrote:
>>

[snip to avoid NNTP message about excessive quoted lines]

>
> There is almost no point in running TestDisk at this point.
>
> The only point in running TestDisk (from Windows), is if Disk 3 is
> being reported as a 2.2TB disk. You recently ran HDTune, and the bad
> block scan only spanned 800GB. Which means, Windows is preventing any
> operation above the 800GB point. If you cannot get above the 800GB
> point, "no way or no how", then TestDisk will run into the same limit.
> If it can't see the whole disk, it's useless.
>
> TestDisk can also be run from linux, from a Linux LiveCD. You
> may even find a copy of TestDisk is standard on the LiveCD (doesn't
> require a trip to Synaptic Package Manager, to download a copy).
> So you can try that if you want.
>

Is TestDisk likely to report any differently from Linux than from
Windows? That's probably a dumb question. I should do it and see for
myself. As a different OS, it may well produce very different results.

How about no data fields at all?

When I look at the two smaller drives, I get a bunch of lines of a
report with numbers(Raw Error Read Rate, Spin Up Time, etc. etc.); some
of the lines are highlighted in yellow (Reallocated Sector Count, Spin
Retry Count, and [B8] Unknown attribute). The same three attributes are
highlighted on the other small driver.

But when I look at the big drive, there are no attributes at all. The
main part of the window is completely blank. Beneath the main listing,
it says "Power On Time: 27382" and "Health Status: Ok". That's all. It
looks as if things are so muddled there that HD Tune can't find any of
the relevant information.

I'm going on to your latest post now....

--
Rhino

Rhino

unread,

May 25, 2013, 1:44:36 PM5/25/13

to

On 2013-05-25 12:50 PM, Paul wrote:
> Rhino wrote:
>> On 2013-05-25 10:42, Paul wrote:
>>> Rhino wrote:
>>>> On 2013-05-24 19:54, Paul wrote:
>>>>> Rhino wrote:
>>>>>

[snip to prevent NNTP errors]

Aaargggh! As if I don't have enough other problems, now Thunderbird is
refusing to post even the much snipped version of this reply due to
"Multipost", even though I am posting to only a single newsgroup....

>
> If the drive capacity is listed as 800GB, and TestDisk has even given
> you a warning, there's no point going much further with TestDisk under
> Windows.
>
> It's of minor interest, to use TestDisk on a platform that reports
> Disk 3 as being 2.2TB, as then, you can view files. I.e. The partition
> still doesn't mount (because it is broken), but TestDisk can attempt
> to use what information is there, to display some of the files. Copying
> files from the damaged section, might still be impossible in any case.
>
> Your priorities:
>
> 1) Come up with an environment, where the Acronis driver can turn
> Disk 3, into Disk 3 and Disk 4, and give you back access to the data.
>

What are you proposing as the best way for me to do that?
a) Uninstall the Acronis application, remove the drivers (if that is
possible), and remove the "bits and pieces" (registry entries, services,
etc.)
b) Revert the computer to Windows XP SP2 via the disc image I have for
that, then upgrade to SP3 via the Microsoft Update process, then attempt
to reinstall DiscWizard or Acronis OEM and hope the big drive recognizes
okay?

Shouldn't I boot up with Linux FIRST and try to see/rescue as much of my
data as possible in case any of this goes wrong?

Have we ruled out the idea of changes in the BIOS to make the big drive
recognize better? Have we ruled out simply swapping the SATA connectors
around to see if the big drive is recognized better?

> 2) There is a minor possibility the status of the right-most partition
> (broken), might change during step 1. But if not, you *still* have to
> complete step 1, in order for *any* utility to recover the files
> off the Disk 3 right-most partition. As long as that disk is stuck
> at reporting 800GB, you're not going to get any access to the files.
>
> HTH,
> Paul

--
Rhino

--
Rhino

Paul

unread,

May 25, 2013, 4:18:28 PM5/25/13

to

I'm sorry I didn't reply about the SATA connectors first.

I do not expect any difference in symptoms if you change
SATA connectors. The same discovery process should take place,
no matter which connector is used. Since you *are* able to
communicate with the drive, I don't see how things could change.

Changing SATA connectors might be important, if there was no
response from the drive at all.

*******

You could back up C:, then install to C: again using your WinXP CD.
There are free backup utilities. You will need at least one blank
CD to use this one, Macrium Reflect Free (download near bottom).

http://www.macrium.com/reflectfree.aspx

Install the app, fire it up, go to the "Other Tasks" tab and
"Create Rescue Media". Use the internal Linux CD they have
in the application - don't bother with the additional 1GB
WAIK kit download to make a WinPE disc. Test that the disc
boots OK, before going further.

Macrium uses VSS to make backup copies. It can back up C: without
a Ghost style "reboot". You only need as much space for a backup,
as is occupied by files on C:. Place the resulting image on some
partition that you know will survive what you're about to do.

Now, once you have backup in hand, you can blow away C: and
install from your WinXP Sp2 disc. I wouldn't bother with a
zillion Windows Updates or the SP3 service pack. You're only going
to need this version of the OS, for the duration of your data
recovery efforts. You want your new motherboard CD handy as well,
because you're going to need the drivers from it. You do the
drivers, after the basic install is in place. And with respect
to drivers, you want to check what mode the NVidia storage ports
are in, in order to determine if you need to press F6 or not.
If the SATA ports are in "IDE" mode, you would not need any
special procedure. Whereas AHCI or RAID, would require pressing
F6, and consulting the manual on how to make a floppy with the
necessary txtsetup.oem and other driver files.

When doing the Windows install, only the target disk should be
cabled up. That's to avoid accidents. I've installed a few
different OSes over the years here, and have had enough "surprises",
to use that rule. I leave only the optical drive, and one disk
drive connected, while installing an OS.

Anyway, once completed, connect up the other disks again,
find your Acronis software, whatever one is up to date,
and install that. You claimed in a post some time ago, to be
using a more recent Acronis than I've got.

*******

You can try booting the Linux LiveCD if you want, and
trying TestDisk from there.

sudo fdisk /dev/sda
p <--- print current disk details
q <--- quit, don't make any changes

sudo fdisk /dev/sdb <--- check the second disk
...

sudo testdisk <--- sudo makes you the admin...

Don't save any new MBR computed with TestDisk. You
don't need it right now. You can use TestDisk, to
examine files on the right-most partition of the
big disk. I don't think Linux can do anything
about the virtual thing that Acronis makes from
the "hidden 746GB" thing above 2.2TB.

If you need to quit TestDisk, you should be able
to abort at any time by pressing <control-C>.

Linux doesn't have a CHKDSK. If you use the CHKDSK
there, it sets a bit to cause a CHKDSK run, the next
time Windows is booted. I don't think you can repair
NTFS from there.

I won't write anything about data recovery yet, until
the Acronis is working again.

Later, you'll be booting from the Macrium Reflect
recovery CD, to restore the bootable C: to the drive
it was on.

I don't use exactly that method, when I put back an old
WinXP version. I copy the entire partition with "dd", and
do that from Linux. But it's a bit complicated to explain
all the manual details involved. I'm suggesting Macrium
in this case, because the backup takes less space, and
it's got a decent GUI - less screwing around.

HTH,
Paul

Rhino

unread,

May 26, 2013, 1:47:46 PM5/26/13

to

No worries.

> I do not expect any difference in symptoms if you change
> SATA connectors. The same discovery process should take place,
> no matter which connector is used. Since you *are* able to
> communicate with the drive, I don't see how things could change.
>
> Changing SATA connectors might be important, if there was no
> response from the drive at all.
>

Fair enough. I just wanted to be sure we didn't skip a possibly very
easy fix in favour of a much harder one.

> *******
>
> You could back up C:, then install to C: again using your WinXP CD.

Do you mean the original XP without service packs or XP with SP2? I've
misplaced the XP with SP2 CD but I can see the image on my hard drive; I
can burn that to another blank CD-R if necessary.

I thought backing up the C: was going to get interesting. There are 150
GB of data, mostly email, but also lots of program settings and other
odds and ends. I wasn't sure where I was going to put all the data. I
have a spool of DVD-RWs but not enough for that much data. But I googled
and realized that I could copy from one computer to the other over the
network easily enough. I've already started that process for the C:
drive. The laptop is wireless though so 150 GB may end up being a LONG
process. Well, if it looks bad, I'll bring the laptop upstairs and
connect it with ethernet; that should speed things up. It's still doing
discovery and has already found 160,000 items....

It looks like it wants to copy; the laptop is complaining that it needs
permission from the (desktop) computer's administrator to make changes
to a folder even though I am only copying it to an archive with no
intent of changing it. I've gone into the drive list under My Computers,
right-clicked on C:, and checked the "Allow Network users to change my
files" to get past the problem the laptop is having, even though "Share
this folder on the network" is already checked. It has been slogging
away for several minutes "setting folder permissions". I'm guessing that
it is updating the file permissions on each of the 160,000+ files on the
C: drive individually. What a bunch of nonsense! A substantial amount of
work and a corresponding delay in what I'm trying to do yet this
shouldn't be necessary to simply copy a file from one computer to the
other as I see it....

I'll wait for this to finish, then continue on with Macrium. (I've
already downloaded the Macrium program; I just haven't run it yet.)

> There are free backup utilities. You will need at least one blank
> CD to use this one, Macrium Reflect Free (download near bottom).
>
> http://www.macrium.com/reflectfree.aspx
>

> Install the app, fire it up, go to the "Other Tasks" tab and
> "Create Rescue Media". Use the internal Linux CD they have
> in the application - don't bother with the additional 1GB
> WAIK kit download to make a WinPE disc. Test that the disc
> boots OK, before going further.
>

Okay, I installed Macrium and I just tried to create a Linux rescue disk
with it. I was tempted to choose both "compatibility mode" and "debug
mode" but, since you didn't say I should choose those, I left them
unchecked. I clicked Finish (the Next option was greyed out) and it
started trying to create my rescue disk - and promptly konked out with
"An error occurred creating the Rescue Media". I had pointed it to my
burner (drive F:) and there was a virgin CD-R in the drive. (At least I
assumed it was completely unused and now it's too late to check.)

Now what? Try with another blank CD-R? Try with a different backup
program? I have Nero 10 on the computer and I'm pretty sure it has a
backup feature.

> Macrium uses VSS to make backup copies. It can back up C: without
> a Ghost style "reboot". You only need as much space for a backup,
> as is occupied by files on C:. Place the resulting image on some
> partition that you know will survive what you're about to do.
>

Just to be clear then, once we have the rescue disk created, I'll still
want to backup C: (I'll use D: since it has plenty of room). Then, we
reinstall XP SP2 and try to rescue the data from the 3 TB drive. Right?

The thing that's confusing me is that I distinctly remember reading on
the box of the 3 TB drive that I had to have SP3 before I could even
begin to use that drive and even then, only after I'd installed
DiscWizard and its friends. Therefore, don't I have to upgrade SP2 to
SP3 before I can hope to see the data on the 3TB drive and copy it off
to somewhere safe?

Should I at least try Unbuntu to see if I can access all or
most of the data on the 3 TB drive before I attempt to clobber Windows?

--
Rhino

Paul

unread,

May 26, 2013, 3:53:19 PM5/26/13

to

Macrium dialog, to create the rescue disc, has a "Create ISO"
option. As shown here. In other words, you can have Macrium
dump an image of the CD, then use whatever tool you want
to actually burn the CD.

http://imageshack.us/a/img560/7969/macriumrescue.gif

Once you have the iso9660 file, you can use Imgburn if you want,
Nero, or whatever tool you normally use for converting .iso
files, into a bootable CD. (Do not "drag and drop" the ISO
onto the volume. ISO9660 files must be converted by a
burner program.) When you're finished, check that
the CD actually boots. Don't nuke your C:, until you have
some idea the recovery is going to work.

*******

You can make a slipstreamed installer CD. I did that for
my Win2K CD, converting it from Win2K SP2 to Win2K SP4.
At the time, Autostreamer did the job. For WinXP, you
can try NLite.

Notice, how it has an "Integrate a Service Pack" item.

http://www.nliteos.com/guide/part1.html

You can get SP3 as a redistributable file.

http://www.microsoft.com/en-us/download/details.aspx?id=24

WindowsXP-KB936929-SP3-x86-ENU.exe

331,805,736 bytes

So those are some materials for making your own installer disc.

*******

You have all the materials. Do whatever you have time for.

Making all portions of the disk visible again,
seems to me to be pretty important. Now, I wouldn't have
to worry about that, because I have a recipe for extracting
sections of the 3TB drive (from Linux, and "dd"), and restore
the image to a smaller disc (like, my 2TB drive). But if
you're not interested in doing that, I see nothing wrong
with backing up C:, installing fresh WinXP, installing
Acronis, just so you can view the 3TB drive. Then, you're
in a better position to attempt data recovery. Where you
put the files, all depends on "space in hand".

When Macrium makes the backup, I think it has a compression
option. If you're desperate for space for C:, you can tick
that box. The backup will likely run slower. But you'll need
less room on D: for the image.

As for permission problems, sometimes when I do backups, I
"stage" them on the local disk. Then, I handle the network
copy as a separate step. For example, this is how I do my
laptop on occasion. Backup C: to D:. FTP the file on D:
over to a computer on my LAN, then erase the copy on D:.
That's hard on the hard drive (when backup runs from C: to
D:), but it's marginally faster than using an external
USB2 enclosure. My network can do around 100MB/sec, while
USB2 disk drive is 30MB/sec.

+--------+--------------+ +----
| C: | D: for data | ----- FTP to computer ---> |
+--------+--------------+ +----
| ^
v |
Backup, to D: ---+

If I need to reverse that process, I'll need my Linux LiveCD
to set up a means to copy the file back to D:. Or, load the
image file onto a USB enclosure drive, and plug into the laptop.
Lots of options. I've even used RCP (remote copy) in the
Linux environment. The security when doing stuff like
that, is a damn nuisance.

http://en.wikipedia.org/wiki/Rcp_(Unix)

*******

One advantage of using the latest Service Pack for
your temporary install, is not running into the 137GB
issue on IDE drives. I think one of the Service Packs
(SP1a?) still had that problem.

(This is irrelevant now, but at one time used to come up
with users trying to reinstall. This problem arose on
my Win2K SP2 installer CD, and prevented access above
137GB on my crusty old IDE drives.)

http://support.microsoft.com/default.aspx?scid=kb;en-us;q303013

HTH,
Paul

Rhino

unread,

May 26, 2013, 7:41:08 PM5/26/13

to

Let's start with some good news!!

I made the ISO image and burned the image to CD with Nero. The image was
made in just seconds and the burning of the disk was quick too. Then I
changed my (desktop) BIOS to boot from CD first. I booted with the
rescue disk in the drive. There were some error messages but nothing too
severe apparently because it got into the GUI. I was asked which image I
wished to restore and got taking to a file directory. I saw two out of
my three logical drives (F: and G:) and both seemed to be completely
intact; all the directories I expected were present on both; the
subdirectories also appeared to be exactly what they should have been in
each case. I couldn't see any files in the subdirectories but I think
this GUI is only showing files with an extension appropriate for a
backup image so I'm not that worried yet.

A couple of odd things: The F: drive was displayed as if it were
actually the G: drive and vice versa. The E: drive was completely wrong
and contained nothing remotely like it should have. There are just two
directories on it, boot and EFI, and boot contains two folders, grub and
grub2. EFI contains no folders. I don't know what's going on there but
I'm guessing that it can't see my E: drive at all and is showing me the
contents of the rescue disk CD. Yeah, I'm pretty sure that is it because
it shows an H: drive with exactly the same layout as the E:.

I'm not sure what to make of that but the apparent survival of my F: and
G: drives (the two most important drives making up the 3 TB drive!) does
my heart good!

Now, let's figure something out. I think something fell through the
cracks because I don't have a backup of C: yet and I'm not clear how I
was supposed to do that.

Am I right in believing that I should use Macrium for both purposes?
First make the rescue disk, then use Macrium again (from Windows or from
the rescue disk??) to backup C: to somewhere suitable (D: should have
enough room although I wouldn't be opposed to backing up C: to my laptop
either; then I can copy from the big hard drive to D: (and C:?) on my
desktop. I'm still not sure how big the backups are going to be; I'm not
even sure how big the files are, let alone how big they will be if put
into a backup since I'm not sure if the backup is doing compression.

> *******
>
> You can make a slipstreamed installer CD. I did that for
> my Win2K CD, converting it from Win2K SP2 to Win2K SP4.
> At the time, Autostreamer did the job. For WinXP, you
> can try NLite.
>
> Notice, how it has an "Integrate a Service Pack" item.
>
> http://www.nliteos.com/guide/part1.html
>
> You can get SP3 as a redistributable file.
>
> http://www.microsoft.com/en-us/download/details.aspx?id=24
>
> WindowsXP-KB936929-SP3-x86-ENU.exe
>
> 331,805,736 bytes
>
> So those are some materials for making your own installer disc.
>

My SP2 disk was made by slipstreaming XP (without service packs) and
SP2. I think I used Autostreamer; that sounds familiar. I seem to
remember that being a fairly painless task and not as confusing as some
of these things sometimes are so I'd be inclined to use it again over
NLite or other things unless there's a problem with that approach.

So we're agreed that I need to install XP with SP3, not XP with SP2,
once I scrub the C: disk and reinstall Windows?

> *******
>
> You have all the materials. Do whatever you have time for.
>
> Making all portions of the disk visible again,
> seems to me to be pretty important. Now, I wouldn't have
> to worry about that, because I have a recipe for extracting
> sections of the 3TB drive (from Linux, and "dd"), and restore
> the image to a smaller disc (like, my 2TB drive). But if
> you're not interested in doing that, I see nothing wrong
> with backing up C:, installing fresh WinXP, installing
> Acronis, just so you can view the 3TB drive. Then, you're
> in a better position to attempt data recovery. Where you
> put the files, all depends on "space in hand".
>

I'm going to research my options on that while I'm awaiting your reply
to this post. I think the computer store is closed by now on a Sunday so
I can't determine if they do drive rentals but I can at least see if
someone in town does, assuming they mention it on their web page.... I
also want to see if I can find sufficient space online for free although
I'm not optimistic; I think I'll get charged for anything more than 5 or
10 GB.

I mostly use FileZilla these days; I think they have a server to go with
their client. Maybe I'll try installing that and see how I like it.

> *******
>
> One advantage of using the latest Service Pack for
> your temporary install, is not running into the 137GB
> issue on IDE drives. I think one of the Service Packs
> (SP1a?) still had that problem.
>
> (This is irrelevant now, but at one time used to come up
> with users trying to reinstall. This problem arose on
> my Win2K SP2 installer CD, and prevented access above
> 137GB on my crusty old IDE drives.)
>
> http://support.microsoft.com/default.aspx?scid=kb;en-us;q303013
>

I'm getting confused about timing. Now that I have the rescue disk
created - but no backups of anything yet - am I installing XP with SP3
so that I can do the backups? I thought you'd suggesting taking the
backup of C: first, then reinstalling Windows with SP but NO
Acronis/DiscWizard, then install a new Acronis/DiscWizard so that we
could see ALL of the big drive, then copy off as much of the big drive
as we can. All of that to be followed by upgrade to Win7 or Win8 maybe
since they will have no trouble seeing all of the 3 TB drive via GPT and
no need for this Acronis driver nonsense.

Have I got that about right?

For what it's worth, I'm not clear on how I'm going to take the backup
of C. Am I booting back into my current Windows XP SP3 (before anything
has deleted) and using Macrium to back it up and then copy it to D: or
perhaps to the laptop?

I've downloaded FileZilla server on the laptop but haven't installed it.
I'd put the server on the computer that's going to receive the files,
right? I haven't done a computer-to-computer-over-the-network FTP in
quite a while and have never done it on my own network that I can recall....

--
Rhino

unread,

May 28, 2013, 7:12:56 PM5/28/13

to

I haven't updated this thread in a couple of days so I thought I'd
drop in with some news.

I've used that Ubuntu disk in LiveCD mode and made some major headway
with it. As I'd hoped, booting from the Ubuntu CD brought me to a menu
offering to let me install it or just pretend that Linux was installed;
I chose the latter and was rewarded with full access to my F: and G:
drives. I've been working on copying as much of the data as possible
to my D: drive. It's clearly not all going to fit. I've done some
house-cleaning and gotten rid of data that I really don't think I'll
need again. It still won't fit into the space that I have. I'll save
the most important stuff on the D:, on DVD-RWs and on my laptop and
see how important the remaining data is; with a bit of luck, it will
be things that I can live without. Otherwise, I may be heading to the
store for another drive, the smallest I can get away with.

I've also discovered that I have sound in Ubuntu! I've tried playing some
videos in Ubuntu and I can hear them just fine. No more lip-reading to make
out what's going on ;-) So we know the board is okay with respect to sound;
it's just the drivers that aren't letting Windows access the audio.

I still haven't seen hide nor hair of my E: drive, the remaining partition
of the 3 TB drive. But I haven't tried all of your suggestions yet; I'll get
to that after I've salvaged as much of the F: and G: drives as I can.
Those are the most critical files in any case.

By the way, I should clarify that I have not cleared off the C: drive
yet or even slipstreamed XP and SP3. I'm using the Ubuntu disk to do as
much as I can before I mess with Windows.

I'll update further when I have more news. Thanks for all your help so
far Paul! There's still lots to do but there is positive stuff happening
and I'm making real headway toward a full resolution.

--
Rhino

Paul

unread,

May 28, 2013, 11:25:31 PM5/28/13

to

You can do it all from Linux. Using the example method I posted, involving
"dd" to dump a copy of the upper portion of the disk, and store that
746GB block of sectors as a single file in one of the (now empty) lower
partitions. Then, loopback mount it (-o loop) and it will show up
as an NTFS partition. Then you can copy the files, either right into
one of the lower partitions, or where ever you want.

When I did it, I used a sparse file technique, because I knew my test
partition was mostly empty. That probably would not be a profitable
option for your.

My example method started about half way down this page.

"Now, back to the data recovery tutorial"

http://groups.google.com/group/alt.comp.periphs.mainboard.asus/msg/4556536356a2fa2c?hl=en-GB&dmode=source

| 746GB | 1454GB | 746GB |
^ copy_Sectors
| as a file and
| store it here
| |
+-----------+
Then, loopback mount the
746GB file. You'll have maybe
500GB left in the partition,
to put recovered files.

I used the "hexdump" program, to probe the
disk, and make sure I was pointing at the correct
starting location. It only works, as long as I've
grabbed precisely the correct starting sector.

As far as I know, in Linux, transporting the sectors using "dd", is
the only way to get at what is up there.

If Acronis made a Linux driver, this would be so much easier.

*******

The "reinstall WinXP, install Acronis, recover upper partition"
method is still available, if you don't want to do that. The
"reinstall WinXP" method, might be slightly quicker to complete.
It would take quite a few hours, to "dd" the 746GB set of
sectors.

It's really an ugly problem, done either way. And wouldn't
be necessary, if Acronis made better installers, that
could clean up after themselves. I've had a fair amount
of trouble here so far, getting the partitions on the 3TB
drive to show up.

Paul

Paul

unread,

May 29, 2013, 1:36:57 AM5/29/13

to

Some good news. I found a way to do it from Linux, without
copying anything. So this will give you access to the top
files, instantly.

It turns out, the mount command accepts an offset, so you
can basically manually step the mounter out to that partition
way up in 3TB country.

Let's say the 3TB disk is /dev/sdc (that's where it is right now,
because my disk setup is slightly different than before).

I figure out which is which, by doing

sudo fdisk /dev/sdc
p <--- print parameters for the disk
q <--- quit without changing anything

The 3TB disk, will have some partitions with fairly large LBA values.
And /dev/sdc matches the description.

Once I've found the disk (sda, sdb, sdc, etc), I use hexdump
to inspect for an NTFS header. This assumes the Acronis Extended
Capacity Manager portion above 3TB, has the one partition. I'm
just going from memory here, as to what this number represents.
This number should also work for you, but verify with the "hexdump"
command. I've included a couple lines, of the hexdump output.

Now, here, 0x20000140000 = 2.2TB + 262144 bytes gap + Acronis 1MB offset

sudo hexdump -C -s 0x20000140000 -n 512 /dev/sdc
20000140000 eb 52 90 4e 54 46 53 20 20 20 20 00 02 08 00 00 |.R.NTFS .....|
20000140010 00 00 00 00 00 f8 00 00 3f 00 ff 00 00 08 00 00 |........?.......|

What I'm looking for there, is evidence I've found the beginning of the file
system, just above 2.2TB. And seeing the letters NTFS like that, in the
result, proves I'm in the right place.

I convert the hex number to decimal, and verify I get the same result

sudo hexdump -C -s 2199024566272 -n 512 /dev/sdc

OK, now make an empty directory. This is a mount point, and the
file system will mount over top of this and your files will be
found there. Since my Windows partition was called POKEY, that's
what I called it in Linux too. (Normally, the Linux automounter
would figure out the label on its own, but we're doing this
the old-fashioned manual way.)

sudo mkdir /media/POKEY

And this is the command I learned tonight. Since I have proof the
sector at 2199024566272 is the beginning of the top-most partition,
I attempt to mount it. If this isn't the beginning of the partition,
I'll get a snotty error message for my trouble, and nothing will mount.
I tried this, and it works. I can see the top partition now!

sudo mount -t ntfs -o offset=2199024566272 /dev/sdc /media/POKEY

If you want to mount the partition read-only, so nothing bad can
happen to the top partition, that would look like this. You
cannot delete files in /media/POKEY, if you run it this way.

sudo mount -t ntfs -o ro,offset=2199024566272 /dev/sdc /media/POKEY

When you're finished copying files off it, this unmounts it.

sudo umount /media/POKEY

You can remove the now-empty mount point as well, but since
you're using a Linux LiveCD, it disappears as soon as you
shutdown. I need to remove the directory here, because my
copy of Ubuntu is running from a USB stick (persistent
file method).

Have fun,
Paul