The second report addresses Intel's new Xeon 5100 processors, also known as Woodcrest. The Inquirer claims the chips were the subject of trials in "one of the US government departments," but that they were rejected early on due to similar hiccups. Allegedly, "severe problems" occurred throughout benchmarking and day-to-day use when the Xeon machines were outfitted with IBM ServeRAID controllers and set up to run a RAID 1E array.
Interestingly, on the same day as The Inquirer's Woodcrest story, HKEPC said Intel was hinting that final versions of the Core 2 Duo would feature a new B-2 stepping. HKEPC's report doesn't delve too deeply in explanations, but it does state, "As for the previous version, a problem was found to make the system full loaded. It’s only solved in the new stepping." Could this report refer to the alleged CPU utilization problems with RAID configurations?