Intel ICH6RW Recall:
While we are unsure if it will be classified as a "recall", there are
certainly going to be many new Intel equipped motherboards pulled back from
the channel as reported here on Thursday. The ICH6RW southbridge chipset
supports the new motherboard 925/915 chipsets from Intel and were launched
earlier this week. From information obtained today, it looks as though Intel
is using a yet to be announced fabrication process on their ICH6RW chipset
that is running them into some trouble due to leaking current. It was related
to us that the leaking current could weaken the onboard battery prematurely
causing BIOS resets, which would of course dump all the configuration
settings in the BIOS back to default. With current RAID setups and such
supported in the BIOS, a loss of BIOS settings could cause a no-boot
situation as the RAID configuration is lost. Even though this could be easily
fixed for the short term, it is surely an issue that Intel is currently
An estimated number of 100,000 units are to be impacted with this issue.
Retailers and distributors are currently pulling many brands of 925/915
motherboards out of the channel. Our performance testing of 925/915 can be
found here . We are still waiting on an official statement from Intel.
**UPDATE** 6/24/04 11:08pm: It has been confirmed that this problem is NOT an
overall problem or defect with the new ICH6 southbridge. It is rather being
classified as a "Fab Excursion." It seems as though a layer of thin film was
not properly removed from the ICH6 chipsets causing the current leakage.
Intel has identified which chipsets have been affected and will be removing
those from the marketplace. An official statement will be forthcoming, but we
have been told that all 925/915 boards in the channel are being pulled for
the shelves for refund. Seeing how no PCI-Express video cards are for sale, I
doubt very many of these motherboards have been sold.
**UPDATE** 6/25/04 11:39am: Intel did give us a statement to pass along to
you. This of course verifies most of our previously reported information.
Intel has identified a manufacturing excursion that affects a limited number
of I/O controllers that are part of the 915/925 chipsets. This excursion --
caused by incomplete removal of thin film on the die pad area – makes the
controller susceptible to excessive leakage in the Real Time Clock (RTC)
circuitry. If you were to get one of these controllers, the likely results
would be a failure to boot, a system hang, or other anomalous system
behavior. We are checking 915/925 chipsets that were shipped prior to, and as
of, the day of launch (June 21, 2004). We have asked for affected shipments
of this product to be returned to Intel for inspection or replacement. We
believe almost none (a few thousand units) of the impacted product has
reached the end customer. We are working with our customers to find the
impacted manufacturing lots and replace them with good product. All product
shipping now is unaffected—this is not an errata or “bug”.
Word has been spreading that there are some series issues with Intel's latest
and greatest chipsets that could cause a system to fail boot up or freeze.
The I/O hub controls some of Grantsdale's new features, including the
circuitry to turn a PC using the chip set into a wireless access point, a
capability Intel has delayed. Intel executives have called the chip set "the
most significant platform in 12 years."
The glitch is somewhat complicated. Normally, when a chip is being
fabricated, an insulating layer of film is deposited to electrically isolate
the chip. That film is normally removed from the die pad area, where the chip
interfaces with the pins that connect it to the outside world.
Though Intel claims very little of the affected chipsets made it out.
"Almost none of it is in the end-user community," he said of the affected
chip sets. All of the PCs that are shipping now are free from the glitch, he