Intel Identifies Chipset Design Error, Implementing Solution

[duckula]

Intel Identifies Chipset Design Error, Implementing Solution

SANTA CLARA, Calif., Jan. 31, 2011 - As part of ongoing quality assurance, Intel Corporation has discovered a design issue in a recently released support chip, the Intel® 6 Series, code-named Cougar Point, and has implemented a silicon fix. In some cases, the Serial-ATA (SATA) ports within the chipsets may degrade over time, potentially impacting the performance or functionality of SATA-linked devices such as hard disk drives and DVD-drives. The chipset is utilized in PCs with Intel's latest Second Generation Intel Core processors, code-named Sandy Bridge. Intel has stopped shipment of the affected support chip from its factories. Intel has corrected the design issue, and has begun manufacturing a new version of the support chip which will resolve the issue. The Sandy Bridge microprocessor is unaffected and no other products are affected by this issue.

Read more here (http://newsroom.intel.com/community/intel_newsroom/blog/2011/01/31/intel-identifies-chipset-design-error-implementing-solution?cid=rss-258152-c1-264102).

[allikat]

If you use Sandy Bridge (socket 1155) then your solution is to run all hard-drives and any SSD devices on your SATAIII ports or those controlled by extra controllers until such time as your motherboard manufacturer sorts out the process for the recall.
The problem only affects the chipset controlled SATAII ports, all others are safe to use.

[rafal_iB_PL]

So, barring SATA3 ports, Jmicron would be the way to go now?

[McSteel]

While this is a real issue, it's a bit overblown, IMHO.
It's an issue with one of the PLL comparator transistors, whose gate layer is too thin for the bias voltage that Intel applied. The resulting lower electric field that blocks the leakage current is weaker than the design calls for, and the effect of the increased leakage current is cumulative, slowly degrading the oxide layer of the gate over time. In typical usage scenario, i.e. 2 drives connected with about 2-3 GB per day of total data transfer, the faulty transistor is expected to give out in about 3 years.
That's akin to a MTBF rating, so one could see the transistor fail in < 1.5 years, or, conversely in > 4.5 years. The danger of data loss is minimal, and the maximum loss would be equal to the size of write buffer, which is, in most cases, negligible.

[kidsafe]

Yeah, switched my SSD and /Users drive tot he SATA 3.0 Intel SATA ports. Optical drive and shadow copies HDD on the Marvell controller. At first I accidentally put my SSD on the Marvell controller and wow, it's a PoS...

[McSteel]

You can expect the new revision of Intel's SATA II ports, with the thicker gate used, to be a bit slower than they are now, since thicker gate means the transistor requires a stronger signal to switch on, which means either higher voltage (thus negating the fix) or lower transfer rate.

[rafal_iB_PL]

Which is a non issue, as HDDs are far from SATA2 max bandwidth in practical anyway, and SSDs start to use SATA3.

[walterm]

Future think, should all the ports be SATA iii?

[mariush]

Both Gigabyte and Asus offer either return to store to be sent a new version of the motherboards around May or a refund.

My advice, get the refund if you can and wait until june for Z68 chipset ..

[Elledan]

You can expect the new revision of Intel's SATA II ports, with the thicker gate used, to be a bit slower than they are now, since thicker gate means the transistor requires a stronger signal to switch on, which means either higher voltage (thus negating the fix) or lower transfer rate.

Better solution, just disable the transistor in question:
One fix for this type of a problem would be to scale down the voltage applied across the problematic transistor. In this case there’s a much simpler option. The source of the problem is actually not even a key part of the 6-series chipset design, it’s remnant of an earlier design that’s no longer needed. In our Sandy Bridge review I pointed out the fair amount of design reuse that was done in creating the 6-series chipset. The solution Intel has devised is to simply remove voltage to the transistor. The chip is functionally no different, but by permanently disabling the transistor the problem will never arise.
http://www.anandtech.com/show/4143/the-source-of-intels-cougar-point-sata-bug

[Cprossu]

What's funny to me is that the review samples (non B1 stepping) were without that bug. It must have been removed (even if it was unintentional) and copy pasted back into the design!

This and other issues like it is why I usually wait for a platform to stabilize before jumping onboard. We'll see if they missed any real deal breakers down the line and I sure hope the laptop industry didn't use the sata3g's (in my younger years I'd have asked "Why would they?!" but I never ask why anymore), otherwise this could get out of control like the nvidia mcp's did.

[McSteel]

Better solution, just disable the transistor in question:

http://www.anandtech.com/show/4143/the-source-of-intels-cougar-point-sata-bug

So, what, they just left it in (har har), without any need for it whatsoever?
That's like appointing a traffic police officer to direct traffic on a straight one-way street with no intersections...

[Elledan]

So, what, they just left it in (har har), without any need for it whatsoever?
That's like appointing a traffic police officer to direct traffic on a straight one-way street with no intersections...

Chips these days are designed in Verilog/VHDL languages much like with a software project, just that the bugs are a lot more interesting :) I can imagine a revision control, copy glitch or other (human) error slipping in a bug like this.

For it to happen to Intel is quite an embarrassment. I bet there is a huge reorganizing underway at the responsible department as we speak :)

[Cprossu]

For it to happen to Intel is quite an embarrassment. I bet there is a huge reorganizing underway at the responsible department as we speak
Perhaps they used a orginal socket 5 Pentium 75 to do the division calculations for the power usage of that chipset transistor :lol:

[Elledan]

I wonder whether AMD has got any glitches like these? I recall some stuff with ATi chipsets, but even those were relatively minor. NVidia chipsets were the worst, relatively speaking, with data-corrupting network functions, non-functional NICs and such.

And I don't think Intel will ever live down that Pentium Pro FP glitch :D

[rafal_iB_PL]

Ummm, TLB bug in first Phenoms?

[Cprossu]

or the cold bug in athlon64's due to SOI.

Every CPU or chipset will have bugs of some type do for some reason, the question is can it be taken care of easily in bios/software?

If the answer is no, you've got a problem

[Elledan]

Ummm, TLB bug in first Phenoms?

Ah yes, I recall that one now :) Just shows how easily one forgets mishaps like this, I guess :D

[rafal_iB_PL]

Guys @XS had to go out of their way to replicate it though, it was minor thing in reality blown way out of proportion.

http://www.xtremesystems.org/forums/showthread.php?t=172026