[Update] Intel G33 motherboard's AHCI not detecting hard disks

M

Mr. Man-wai Chang

Original thread:
https://groups.google.com/forum/#!topic/alt.comp.hardware.pc-homebuilt/OayrOg7rlAw

I attached a Western Digital Cavier Blue 500G hard disk
(WD5000AAKS-65TMA0) to the same motherboard. AHCI worked faultlessly.

So the problem was the Seagate 7200.10 250G SATA hard disk (ST3250620AS).

SATA modes should be standard among all hardware devices....I still
could not be sure whether it's a motherboard problem or not.

--
@~@ Remain silent. Nothing from soldiers and magicians is real!
/ v \ Simplicity is Beauty! May the Force and farces be with you!
/( _ )\ (Fedora 19 i686) Linux 3.11.6-200.fc19.i686
^ ^ 18:03:02 up 20:29 0 users load average: 0.00 0.01 0.05
ä¸å€Ÿè²¸! ä¸è©é¨™! ä¸æ´äº¤! ä¸æ‰“交! ä¸æ‰“劫! ä¸è‡ªæ®º! è«‹è€ƒæ…®ç¶œæ´ (CSSA):
http://www.swd.gov.hk/tc/index/site_pubsvc/page_socsecu/sub_addressesa
 
D

Don Phillipson

Original thread:
https://groups.google.com/forum/#!topic/alt.comp.hardware.pc-homebuilt/OayrOg7rlAw

I attached a Western Digital Cavier Blue 500G hard disk
(WD5000AAKS-65TMA0) to the same motherboard. AHCI worked faultlessly.

So the problem was the Seagate 7200.10 250G SATA hard disk (ST3250620AS).

SATA modes should be standard among all hardware devices....I still could
not be sure whether it's a motherboard problem or not.

This query omits whether the OP verified the BIOS is set to enable
both SATA and PATA hard drives.
 
P

Paul

Mr. Man-wai Chang said:
Original thread:
https://groups.google.com/forum/#!topic/alt.comp.hardware.pc-homebuilt/OayrOg7rlAw


I attached a Western Digital Cavier Blue 500G hard disk
(WD5000AAKS-65TMA0) to the same motherboard. AHCI worked faultlessly.

So the problem was the Seagate 7200.10 250G SATA hard disk (ST3250620AS).

SATA modes should be standard among all hardware devices....I still
could not be sure whether it's a motherboard problem or not.

I thought AHCI had fallback behavior, if a device didn't properly
support some part of the AHCI standard.

For example, Raptors (various generations) have TCQ and NCQ. And
NCQ is a part of AHCI. If you plugged an old (TCQ) Raptor into AHCI,
and Windows sees only TCQ, it should just disable the
tagged queuing portion and continue normal (IDE like) operation.
It would still claim to be following the AHCI standard, and
perhaps things like Hot Plug would still work - it just
wouldn't queue up commands for completion out of order.

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

So I don't know what your Seagate drive is doing in this case.

On page two, the Seagate drive does list NCQ support.

http://www.seagate.com/docs/pdf/datasheet/disc/ds_barracuda_7200_10.pdf

The BIOS doesn't need to use AHCI when the system starts up.
AHCI operation starts when the OS driver loads, and control
is transferred to the OS. The BIOS would not be using
command queuing for example, because it isn't necessary.
Command queuing helps, when there is an outstanding
series of requests queued up. And the BIOS tends to be
single threaded (at least the legacy BIOS is, don't know
about UEFI though).

Another avenue to explore, is see what the BIOS tables
report for the hard drive. The BIOS passes a series of
tables, for plug and play usage. While Windows has some
"direct" probes it can use for plug and play (VID/PID or
VEN/DEV), the BIOS tables label things with convenient
generic labels, and that can also trigger the correct
response from Windows detection. An example of that, is
generic "Class Codes" used for hardware detection.

You can go to Device Manager, do Properties on an entry,
use Details tab, then "Hardware Ids" to see more info.

I don't use AHCI on my computer right now, and when
I use Device Manager and look at the Hardware ID, it
reports "PNP0600", which is presumably a value
coming from a BIOS table. The value 197b-2368 is
the VEN/DEV coming from the chipset, and is also
in the hardware ID list.

I don't have enough hardware here, to give you some
representative values for what to expect. I'd have to
switch to AHCI mode on my current computer, and install
an OS to do some testing. Which would take me half the day,
and take me offline :)

Paul
 
F

Flasherly

SATA modes should be standard among all hardware devices....I still
could not be sure whether it's a motherboard problem or not.

Really. 250 and 500G HDs are way back in dating 1 & 2T drives;- we're
on 3T presently. Any, even if an older MB shouldn't have problems in
that storage range. I've also Seagate 250/200G HDs, mixed variously
with WD and Samsung HDs, ranging from 850G to 2T. I don't believe in
the newer HD ranges, though, I've any partitions larger than 1T, which
I generally format from docking stations prior to installing on either
MB, which I believe would be both Gigabyte boards. On one of the
larger HDs, I run it from a $15/US, 2 SATA port, PCI controller. Not
saying everything is perfect, there are a couple of oddities, (for
instance, have a defragger that balks at slow speeds on the PCI
controller, whereas a file copy runs fine at expected speeds), nor do
I particularly necessarily trust those MB and their ages with all the
latest and greatest, especially when looking at 2T drives. I'll also
make one of those older 250/200G drives a main boot disk before using
a bigger HD for other than storage. The boot function, however, may
be superceded with the next 128G SSD I see onsale for $60 (well made
64G SSD's are pretty damn hard to find for $40).

Like I said, a $15 SATA controller is another comparatively cheap
testing route if that MB is picking up some drives while not others.
Not fun when HD storage is in jeopardy, nor a new MB/system build when
a MB rarely, especially an Intel MB, turns to the radical side. Hope
you find a relatively easy fix to your satisfaction.
 
M

Mr. Man-wai Chang

This query omits whether the OP verified the BIOS is set to enable
both SATA and PATA hard drives.

The ICH9 has 3 modes: IDE, AHCI, RAID

--
@~@ Remain silent. Nothing from soldiers and magicians is real!
/ v \ Simplicity is Beauty! May the Force and farces be with you!
/( _ )\ (Fedora 19 i686) Linux 3.11.6-200.fc19.i686
^ ^ 18:51:02 up 19:11 0 users load average: 0.00 0.01 0.05
ä¸å€Ÿè²¸! ä¸è©é¨™! ä¸æ´äº¤! ä¸æ‰“交! ä¸æ‰“劫! ä¸è‡ªæ®º! è«‹è€ƒæ…®ç¶œæ´ (CSSA):
http://www.swd.gov.hk/tc/index/site_pubsvc/page_socsecu/sub_addressesa
 
M

Mr. Man-wai Chang

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

So I don't know what your Seagate drive is doing in this case.
On page two, the Seagate drive does list NCQ support.
http://www.seagate.com/docs/pdf/datasheet/disc/ds_barracuda_7200_10.pdf

The BIOS doesn't need to use AHCI when the system starts up.
AHCI operation starts when the OS driver loads, and control
is transferred to the OS. The BIOS would not be using
command queuing for example, because it isn't necessary.
Command queuing helps, when there is an outstanding
series of requests queued up. And the BIOS tends to be
single threaded (at least the legacy BIOS is, don't know
about UEFI though).
Another avenue to explore, is see what the BIOS tables
report for the hard drive. The BIOS passes a series of
tables, for plug and play usage. While Windows has some
"direct" probes it can use for plug and play (VID/PID or
VEN/DEV), the BIOS tables label things with convenient
generic labels, and that can also trigger the correct
response from Windows detection. An example of that, is
generic "Class Codes" used for hardware detection.

If I set the SATA mode to AHCI and attached the 250G Seagate hard disk
to it, the boot sequence would stop (not sure for how long) at the 250G
hard disk entry.


--
@~@ Remain silent. Nothing from soldiers and magicians is real!
/ v \ Simplicity is Beauty! May the Force and farces be with you!
/( _ )\ (Fedora 19 i686) Linux 3.11.6-200.fc19.i686
^ ^ 18:51:02 up 19:11 0 users load average: 0.00 0.01 0.05
ä¸å€Ÿè²¸! ä¸è©é¨™! ä¸æ´äº¤! ä¸æ‰“交! ä¸æ‰“劫! ä¸è‡ªæ®º! è«‹è€ƒæ…®ç¶œæ´ (CSSA):
http://www.swd.gov.hk/tc/index/site_pubsvc/page_socsecu/sub_addressesa
 
M

Mr. Man-wai Chang

Like I said, a $15 SATA controller is another comparatively cheap
testing route if that MB is picking up some drives while not others.
Not fun when HD storage is in jeopardy, nor a new MB/system build when
a MB rarely, especially an Intel MB, turns to the radical side. Hope
you find a relatively easy fix to your satisfaction.

I attached that same Seagate 250G hard disk to another motherboard, no
AHCI problem at all.

It's time to move forward and sell those old 250G hard disks anyway.

--
@~@ Remain silent. Nothing from soldiers and magicians is real!
/ v \ Simplicity is Beauty! May the Force and farces be with you!
/( _ )\ (Fedora 19 i686) Linux 3.11.6-200.fc19.i686
^ ^ 18:51:02 up 19:11 0 users load average: 0.00 0.01 0.05
ä¸å€Ÿè²¸! ä¸è©é¨™! ä¸æ´äº¤! ä¸æ‰“交! ä¸æ‰“劫! ä¸è‡ªæ®º! è«‹è€ƒæ…®ç¶œæ´ (CSSA):
http://www.swd.gov.hk/tc/index/site_pubsvc/page_socsecu/sub_addressesa
 
F

Flasherly

It's time to move forward and sell those old 250G hard disks anyway.

Strange, I've never had one of my 200/250G Seagates rejected on any
MB. I really like them for beater OS HDs. Divide up some primary
partitions for a few operating systems, a little storage for the rest.
Only problem is it's nice to keep it down to 2 spinning HDs, over
three - less clutter, heat, peak or constant power draws. Unless one
is a SSD, which doesn't use resources though doesn't then give a 250G
much function. I also at one point was angling a SSD boot-installed
arbitrator, defined to boot from the 250. I was having trouble getting
the MD/SSD combo to boot from the SSD's partitions, though eventually
"fixed" it in my own way. Its boot arbitrator is not setup the same
as a spinning platter install, nor am I exactly sure why it works;-
Been awhile and I've half forgotten how I did it, too. (Wasn't
supposed to work, what I attempted though managed, I read, that my MB
support for a SSD wasn't within current specs.)
 
F

Flasherly

that my MB
support for a SSD wasn't within current specs.)

Heh - it's a gigabyte LGA with an the early first dual pentium, rather
than psudo hypthreading CPU intel once was foisting for dual cores. 4
or 5 years easy befor SSDs.

LGA is a nightmare if you ever bend a pin on the MB. I did. I fixed
it, too, but it's like writing on a rice kernel with a fine sewing
needle. Which is what I used, needle and a dispensible razor blade. I
was doing the actual Intel bending under a powerful lighted magnifier
(probably around x20 or x25). Myth of course would have some LGA
masters doing it blindfolded in their sleep.
 
M

Mr. Man-wai Chang

Thank you for sharing your experiences! ;)

--
@~@ Remain silent. Nothing from soldiers and magicians is real!
/ v \ Simplicity is Beauty! May the Force and farces be with you!
/( _ )\ (Fedora 19 i686) Linux 3.11.9-200.fc19.i686
^ ^ 22:06:01 up 2:38 0 users load average: 0.00 0.01 0.05
ä¸å€Ÿè²¸! ä¸è©é¨™! ä¸æ´äº¤! ä¸æ‰“交! ä¸æ‰“劫! ä¸è‡ªæ®º! è«‹è€ƒæ…®ç¶œæ´ (CSSA):
http://www.swd.gov.hk/tc/index/site_pubsvc/page_socsecu/sub_addressesa
 
F

Flasherly

Thank you for sharing your experiences! ;)

FWIW, I'll probably go with AMD pin-sockets from here on out. I'm
just not that wild about Land Grid Arrays now.

(It was a brand new Gigabyte MB, fresh out of the box;- I'd already
found and purchased that Dual Core Pentium on Ebay. There's a plastic
insert MB manufacturers use to protect the LGA female pins, which was
wedged into the socket for some force required, by some unknown
proceedure to remove, whereupon I bent actually several of the CPU
pins in getting it out. My brand new MB then would not boot. Imagine
that, not the after-thought when I eventually concluded I'd have to
fix it myself;- sending it back to the vendor with bent pins obviously
wasn't going to hold water.)
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top