slow ATA 133 disk performance in IBM NetVista 6276 27G PC

  • Thread starter Thread starter rob
  • Start date Start date
R

rob

I have recently replaced the original 10Gb IBM HDD in a IBM NetVista
6276-27G PC with a Maxtor DiamondMax Plus 8 ATA or UDMA 133 40Gb HDD.
It's running, but MUCH slower than the old disk. I am running the
latest BIOS for the machine (version 22) and am a novice in this area.
I see an option on the Maxtor site for an additional PCI IDE ATA 133
adapter but I am not sure this could work as the only disk (is
bootdisk) for a machine. I am also told that a 80 wire IDE cable might
be the solution but I am not convinced that that could cause such a
problem. As I said, I am a novice in THIS area (not on PC's in
general) so would appreciate any detailed advice/instructions. I have
run the program AIDA (http://www.aida32.hu/aida32.php) to show some
diagnostic data on my pc but don't exactly know what to look for and
what it al means. It mentions UDMA 66 and UDMA 100 etc.
Thanks very much for any help!!!
 
rob said:
I have recently replaced the original 10Gb IBM HDD in a IBM NetVista
6276-27G PC with a Maxtor DiamondMax Plus 8 ATA or UDMA 133 40Gb HDD.
It's running, but MUCH slower than the old disk. I am running the
latest BIOS for the machine (version 22) and am a novice in this area.
I see an option on the Maxtor site for an additional PCI IDE ATA 133
adapter but I am not sure this could work as the only disk (is
bootdisk) for a machine. I am also told that a 80 wire IDE cable might
be the solution but I am not convinced that that could cause such a
problem. As I said, I am a novice in THIS area (not on PC's in
general) so would appreciate any detailed advice/instructions. I have
run the program AIDA (http://www.aida32.hu/aida32.php) to show some
diagnostic data on my pc but don't exactly know what to look for and
what it al means. It mentions UDMA 66 and UDMA 100 etc.
Thanks very much for any help!!!

If your data cable is the older 40 connector type, you definitely should
replace it with an 80 connector one. The 40 connector cables have trouble
with the faster data transfer.
 
Folkert Rienstra said:
Nope, they don't.

?

From PCGuide

"Requirement: The 80-conductor cable was first defined with the original
Ultra DMA modes 0, 1 and 2, covering transfer speeds up to 33.3 MB/s. The
cable is considered "optional" for those modes. However, for any Ultra DMA
modes above mode 2, the 80-conductor cable is mandatory.
Detection: Since the cable is mandatory for high-speed modes, the system has
to have some way of knowing it is installed. This is done by having the
/PDIAG:/CBLID signal, carried on pin #34 of the interface, grounded in the
connector that attaches to the motherboard. Since the older 40-conductor
cable would not have this pin grounded, by looking for the grounding on this
pin at startup the host can determine if the 80-conductor cable is
installed. "

From Maxtor's site

"The faster timing requirements of Ultra ATA/66 and above require the use of
an 80-conductor cable. This is necessary for proper operation of UDMA modes
3 and greater. The 80-conductor cable is used with the same connector
configuration as the standard 40-conductor cable. The 40 additional
conductors are used as ground paths and are all connected to the 7 original
ground conductors. These additional ground conductors serve to improve the
overall signal quality (signal-to-noise ratio) of the ATA cable.
80-conductor Ultra ATA cables are also limited to a maximum cable length of
18 inches. "
 
Frank Jelenko said:
?

From PCGuide

"Requirement: The 80-conductor cable was first defined with the original
Ultra DMA modes 0, 1 and 2, covering transfer speeds up to 33.3 MB/s. The
cable is considered "optional" for those modes. However, for any Ultra DMA
modes above mode 2, the 80-conductor cable is mandatory.
Detection: Since the cable is mandatory for high-speed modes, the system has
to have some way of knowing it is installed. This is done by having the
/PDIAG:/CBLID signal, carried on pin #34 of the interface, grounded in the
connector that attaches to the motherboard.
Since the older 40-conductor cable would not have this pin grounded, by looking
for the grounding on this pin at startup the host can determine if the 80-conductor
able is installed. "

And what do you think is the purpose of all this?
Right, that 40-conductor cables don't get to see the faster data rates at all.

So: "Nope, they don't".
 
The 40 connector cables have trouble with the faster data transfer.
Right, that 40-conductor cables don't get to see the faster data rates at all.
So: "Nope, they don't".

Do you mean, "Nope, they don't (in paper)" ? In that case you will be
absolutely correct.

Just had corrupted data on Adaptec TX133 + Maxtor 250 GB HD, didn't have 80
pin cable at hand and figured I don't need faster transfer rates as it's
just for storing movies off the net for streaming around the LAN in my
place. So anyway, tested the transfer and indeed: the files were corrupted!
Checked the UDMA settings, and the Adapted had configured itself using
UDMA-5 !!! I set it to UDMA2, but after reboot it had again switched to
UDMA-5. Needless to say, I got the 80-conductor cable the next morning and
things have worked fine since then.

So: "Yes, they do" (even if not on paper / Rienstra Ideal World), but I
would think that reality is what counts for most of us... the data
corruption possibility is REAL and it shouldn't be taken lightly, if this
avoids some poor soul out there trusting what you say and getting his data
wiped out because he doesn't have back-ups, I will be glad.
 
Do you mean, "Nope, they don't (in paper)" ? In that case you will be
absolutely correct.

Just had corrupted data on Adaptec TX133 + Maxtor 250 GB HD, didn't
have 80 pin cable at hand and figured I don't need faster transfer
rates as it's just for storing movies off the net for streaming around
the LAN in my place. So anyway, tested the transfer and indeed: the
files were corrupted! Checked the UDMA settings, and the Adapted had
configured itself using UDMA-5 !!! I set it to UDMA2, but after reboot
it had again switched to UDMA-5. Needless to say, I got the
80-conductor cable the next morning and things have worked fine since
then.

So: "Yes, they do" (even if not on paper / Rienstra Ideal World), but
I would think that reality is what counts for most of us... the data
corruption possibility is REAL and it shouldn't be taken lightly, if
this avoids some poor soul out there trusting what you say and getting
his data wiped out because he doesn't have back-ups, I will be glad.

I'd squawk at Adaptec about this--their board should not only check what
the drive can do but also check for the correct cable--don't know how
they do this but I've seen enough boards that do it that I know it's
possible and whatever chipset they're using should support it, so they
appear to have a firmware bug that they need to correct.
 
I'd squawk at Adaptec about this--their board should not only check what
the drive can do but also check for the correct cable--don't know how
they do this but I've seen enough boards that do it that I know it's
possible and whatever chipset they're using should support it, so they
appear to have a firmware bug that they need to correct.

Sorry, the part name is:

Promise Ultra133 TX2, connected to Intel GB850. NOT Adaptec. :)
 
Sorry, the part name is:

Promise Ultra133 TX2, connected to Intel GB850. NOT Adaptec. :)

How did I get that it was Adaptec? In that case squawk at Promise,
although much good it will do you, they don't seem to give a damn
whether their product works once you've paid for it. Still a firmware
bug--Intel buys the support module from Promise.
 
John Smith said:
Do you mean, "Nope, they don't (on paper)" ? In that case you will be
absolutely correct.

Just had corrupted data on Adaptec TX133 + Maxtor 250 GB HD, didn't
have 80 pin cable at hand and figured I don't need faster transfer rates as
it's just for storing movies off the net for streaming around the LAN in my
place. So anyway, tested the transfer and indeed: the files were corrupted!
Checked the UDMA settings, and the Adapted had configured itself using
UDMA-5 !!! I set it to UDMA2, but after reboot it had again switched to
UDMA-5. Needless to say, I got the 80-conductor cable the next morning
and things have worked fine since then.

So the card is defective.
So: "Yes, they do" (even if not on paper / Rienstra Ideal World), but I
would think that reality is what counts for most of us... the data
corruption possibility is REAL and it shouldn't be taken lightly, if this
avoids some poor soul out there trusting what you say and getting his data
wiped out because he doesn't have back-ups, I will be glad.

That's a risk that is always there when things go defective.
That's why one always shall have backups.
 
So the card is defective.

So is your claim.
That's a risk that is always there when things go defective.
That's why one always shall have backups.

That's a risk of using a HD in the first place. I ask you, Folkert Rienstra,
how many 'ordinary' people actually DO have backups of everything they have,
even if they should, et cetera et cetera. I wouldn't wager that many.
Atleast of ALL data they've got. If everyone you know, have backups, that's
pretty bizarre.
 
Back
Top