Intel discloses a flaw with its SATA implementation

They say that the 6Gbps ports aren't affected, but I think that's simply
because Intel doesn't produce those ports itself, those are produced by
3rd parties such as Realtek or VIA, etc.

Yousuf Khan

Sandy Bridge chipset flaw to cost Intel $1 billion | ZDNet
"All 6-series “Cougar Point” chipsets, including H67 and P67, are
affected. The issue is a hardware design problem that causes high number
of bit errors eventually device disconnection. The problem only affects
the four 3Gbps SATA ports with the two 6Gbps ports unaffected. According
to Intel, data on connected devices will be safe."
http://www.zdnet.com/blog/hardware/s...-billion/11157

Yousuf Khan wrote:
They say that the 6Gbps ports aren't affected, but I think that's simply
because Intel doesn't produce those ports itself, those are produced by
3rd parties such as Realtek or VIA, etc.

Yousuf Khan

Sandy Bridge chipset flaw to cost Intel $1 billion | ZDNet
"All 6-series ?Cougar Point? chipsets, including H67 and P67, are
affected. The issue is a hardware design problem that causes high number
of bit errors eventually device disconnection. The problem only affects
the four 3Gbps SATA ports with the two 6Gbps ports unaffected. According
to Intel, data on connected devices will be safe."
http://www.zdnet.com/blog/hardware/s...-billion/11157

Ah, finally some detail.

"high number of bit errors" sounds like exceedingly shoddy
design or manufacturing, since this is easy to test for.

Arno

--
Arno Wagner, Dr. sc. techn., Dipl. Inform., CISSP -- Email:
GnuPG: ID: 1E25338F FP: 0C30 5782 9D93 F785 E79C 0296 797F 6B50 1E25 338F
----
Cuddly UI's are the manifestation of wishful thinking. -- Dylan Evans

On Tue, 01 Feb 2011 02:17:21 -0500, Yousuf Khan
put finger to keyboard and composed:

Sandy Bridge chipset flaw to cost Intel $1 billion | ZDNet
"All 6-series “Cougar Point” chipsets, including H67 and P67, are
affected. The issue is a hardware design problem that causes high number
of bit errors eventually device disconnection. The problem only affects
the four 3Gbps SATA ports with the two 6Gbps ports unaffected. According
to Intel, data on connected devices will be safe."
http://www.zdnet.com/blog/hardware/s...-billion/11157



http://www.anandtech.com/show/4143/t...point-sata-bug

"Cougar Point (Intel’s 6-series chipsets: H67/P67) has two sets of
SATA ports: four that support 3Gbps operation, and two that support
6Gbps operation. Each set of ports requires its own PLL source."

"The problem in the chipset was traced back to a transistor in the
3Gbps PLL clocking tree. The aforementioned transistor has a very thin
gate oxide, which allows you to turn it on with a very low voltage.
Unfortunately in this case Intel biased the transistor with too high
of a voltage, resulting in higher than expected leakage current.
Depending on the physical characteristics of the transistor the
leakage current here can increase over time which can ultimately
result in this failure on the 3Gbps ports. The fact that the 3Gbps and
6Gbps circuits have their own independent clocking trees is what
ensures that this problem is limited to only ports 2 - 5 off the
controller."

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

Tue, 01 Feb 2011 02:17:21 -0500, Yousuf Khan wrote:

They say that the 6Gbps ports aren't affected,
but I think that's simply because Intel doesn't produce
those ports itself, those are produced by
3rd parties such as Realtek or VIA, etc.
Yousuf Khan

Sandy Bridge chipset flaw to cost Intel $1 billion
| ZDNet
"All 6-series “Cougar Point” chipsets, including H67
and P67, are affected. The issue is a hardware design
problem that causes high number of bit errors eventually
device disconnection. The problem only affects the four
3Gbps SATA ports with the two 6Gbps ports unaffected.
According to Intel, data on connected devices will be safe."
http://www.zdnet.com/blog/hardware/s...-billion/11157


Not only will it cost Intel big bucks, it'll help to
sustain current AMD cpu prices for xtra months.

I figured that with new "Sandy Bridge" CPUs out
that we'd see a drop in AMD pricing. ...and was
wondering why AMD's CPU prices are static
since XMAS, at least for Phenom II,
which I've been following.

Interesting.

Bill

On 2/3/2011 11:28 PM, willbill wrote:
Not only will it cost Intel big bucks, it'll help to
sustain current AMD cpu prices for xtra months.

I figured that with new "Sandy Bridge" CPUs out
that we'd see a drop in AMD pricing. ...and was
wondering why AMD's CPU prices are static
since XMAS, at least for Phenom II,
which I've been following.

Interesting.

Bill

AMD did announce some price cuts on its six-core Phenom II's recently.

AMD cuts prices on select Phenom II quad-core and six-core processors |
ZDNet
http://www.zdnet.com/blog/computers/...rocessors/5017

Yousuf Khan

Sat, 05 Feb 2011 11:11:18 -0500, Yousuf Khan wrote:

AMD did announce some price cuts on its
six-core Phenom II's recently.

Would you buy a 6-core Phenom II?

If yes, why?

I mean the value (in my current experience) is with
an oem stepping 2 3-core 740 ("Heka" 3.0 GHz) at ~$70.

BTW, have you come across a site/cite that explains
what the current stepping 3 offers over the stepping 2
for Phenom II CPUs?

If yes, please post the link. Thanks.

AMD cuts prices on select Phenom II quad-core and six-core processors |
ZDNet
http://www.zdnet.com/blog/computers/...rocessors/5017

Odds are that if Intel's "Sandy Bridge" hadn't shown
up with problems, then AMD's price drops would
have been forced to be bigger.

FWIW, a few weeks ago I read Anandtech's initial
review on Sandy Bridge and thought: for the likely
price premium that Intel CPUs typically command,
one can still buy a lot of electricity.

Meaning that I'm a long time fan/user of AMD CPUs
and hope that this helps them.

Bill

Sun, 06 Feb 2011 10:25:50 -0600, willbill wrote:

BTW, have you come across a site/cite that explains
what the current stepping 3 offers over the stepping 2
for Phenom II CPUs?

If yes, please post the link. Thanks.

google is still one's friend.

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

http://www.anandtech.com/show/2702/3
(AMD Phenom II X4 940 & 920: A True Return to Competition
dated: 1/8/2009)

http://www.guru3d.com/article/phenom...3-review-test/
(Phenom II X4 965 BE revision C3 review
dated: November 4, 2009)

Bill

On 06/02/2011 11:25 AM, willbill wrote:
Sat, 05 Feb 2011 11:11:18 -0500, Yousuf Khan wrote:

AMD did announce some price cuts on its
six-core Phenom II's recently.

Would you buy a 6-core Phenom II?

If yes, why?

If you have an existing Socket AM2+ or AM3 motherboard, for example. I
currently have a 3-core Phenom II, so going upto only 4 cores is a waste
of time & money, but 6-cores might be on the borderline of worthwhile.
But I really haven't found much that my current 3-core can't handle, so
I probably wouldn't upgrade to a 6-core at this time.

I mean the value (in my current experience) is with
an oem stepping 2 3-core 740 ("Heka" 3.0 GHz) at ~$70.

Possibly, sure. That core probably doesn't have an L3 cache though.

Odds are that if Intel's "Sandy Bridge" hadn't shown
up with problems, then AMD's price drops would
have been forced to be bigger.

Possbily, or the price cuts might have been the same as now but more
widespread across the board. It looks like they only dropped the prices
on the higher-end processors rather than their mainstream processors.

FWIW, a few weeks ago I read Anandtech's initial
review on Sandy Bridge and thought: for the likely
price premium that Intel CPUs typically command,
one can still buy a lot of electricity.

Meaning that I'm a long time fan/user of AMD CPUs
and hope that this helps them.

Their new Atom-killer Fusion processors have caught-on, it looks like.
I'm even seeing some full-sized laptops coming with Zacate processors.
I'd rather not have a netbook processor running in a full-sized laptop,
but they say the video speed is what allows this to be competitive in
this form-factor. Still I prefer somewhat more actual CPU power rather
than GPU power.

Yousuf Khan