Intel’s next-generation Xeon CPUs based on the new Sapphire Rapids architecture have been delayed more than once, but Intel is finally preparing to start selling them to computer companies and end users. according to Advertisement Advertisement Intel’s tweet, “Data center launch event” on January 10 will include Sapphire Rapids processors, the chips have currently met “product release qualifications and the company is beginning to ramp up deployment.”
Also called “Xeon Scalable” or fourth generation Xeon, the Sapphire Rapids CPUs were originally scheduled to launch in late 2021, but by mid-2021 it’s becoming the first quarter of 2022, then “later in the year than originally expected” and now early 2023. We still don’t know when the chips will actually start shipping, but we will get more information in January. These types of delays are relatively common for Intel, which has also struggled to release custom Arc desktop GPUs on time and suffered frequent manufacturing setbacks in the past decade.
In an interview with The Verge last month, Intel CEO Pat Gelsinger spoke about the Sapphire Rapids delays, implicitly blaming previous leadership and noting that future products won’t have the same bumpy offering.
“This project started five years ago, so it’s on board. I can’t reset the methodology for a product that only started five years ago,” Gelsinger said. “[Sapphire Rapids] There was a lot of complexity in it, with three major new systems or interfaces in that design…and there were no backups on any of them. “
Sapphire Rapids is far behind the Alder Lake (12th generation) core processors that have shipped in laptops and desktops for a year or so now, and are built with “Golden Cove”-based CPU cores (as opposed to Alder Lake, Sapphire Rapids uses no low-power cores efficiently) and the same Intel 7 manufacturing process. We’re said to expect up to 60 cores for the datacenter versions for the CPU and up to 56 cores for the workstation versions, at TDPs of up to 350W. But Intel’s delays have made the chips less competitive than they would have been if they launched earlier this year. AMD’s next-generation Epyc CPUs for servers (codenamed Genoa) will offer up to 96 Zen 4 cores per CPU when it launches later this month, while current-generation Threadripper CPUs outperform Already in 64 centers.
Other Sapphire Rapids features include support for DDR5 memory, PCI Express 5.0 connectivity, support for the Compute Express Link (CXL) 1.1 standard, all features that AMD’s Genoa will also support. (These were the “three new major systems” that Gelsinger referred to in his interview.)
Architecturally, one of the most remarkable things about the chip is that it’s Intel’s first foray into chip-based CPUs – each processor is actually made up of multiple silicon molds, linked together with a high-speed internal connection. AMD has used a chiplet-based approach for all Ryzen, Threadripper, and Epyc processors, and it could be a way to improve manufacturing yields; If there is a critical defect in a chiplite mold, you should dispose of much less silicone than you would for a similar defect in one huge processor mold. It also allows for mixing and matching of manufacturing processes, so you can use a sophisticated process for the things you’ll benefit most from (CPU and GPU cores, for example) while using a cheaper, more mature process for other things (I/O and other chip functions).
Intel will rely more on wood-based panels starting with its 14th-generation “Meteor Lake” processors, which will bundle a mixture of “tiles” built using different manufacturing processes together into a single processor.
#Intels #muchdelayed #Xeon #Sapphire #Rapids #CPUs #finally #early