Skip to main content

Intel: Alder Lake Sampling, Sapphire Rapids Samples in Q4

Intel
(Image credit: Intel)

In an update to analysts and investors, Intel has revealed some additional details about its ongoing projects, namely the Alder Laker hybrid processor for client PCs as well as the Sapphire Rapids CPUs for servers. Apparently, Alder Lake is sampling to PC makers, whereas Sapphire Rapids is about to.  

"We are now sampling our 2021 client CPU Alder Lake, and we'll be sampling our 2021 data center CPU, Sapphire Rapids later in the fourth quarter," said Bob Swan, CEO of Intel, during the company's earnings call with analysts and investors. "Both will deliver significant capabilities enabled by our six pillars of innovation including our [10nm] Enhanced SuperFin technology."

Hybrid x86: Intel's Alder Lake is Sampling

Being a rather mysterious processor in Intel's roadmap, Alder Lake promises to bring the concept of heterogeneous multi-core to x86 processors for client PCs. High-end versions of Alder Lake CPUs for desktops are expected to have up to 16 cores. Back in August Intel indeed said that Alder Lake belongs to 'performance' segment of CPUs.  

"We are advancing our hybrid architecture significantly with the focus on performance," said Raja Koduri, chief architect at Intel. "We are working on next generation hardware, guided scheduler, optimize for performance and leveraging all close seamlessly. Alder Lake will not only be great for performance, but it will also be our best performance per Watt architecture." 

Intel plans to use its 4th generation 10 nm process technology — called 10 nm Enhanced SuperFin — to make Alder Lake processors. The process technology was originally designed for datacenter products in mind, which typically means enhancements to power delivery, MEOL and BEOL in semiconductor production terms. Power delivery is crucial for hybrid processors, such as those that use Arm's Big.Little architecture.

Sapphire Rapids Samples in Q4

Releasing first samples of Sapphire Rapids to customers in Q4 2020 is meant to enable them to launch servers based on these CPUs sometimes in late 2021 or early 2021. Intel is supposed to start volume production of its Sapphire Rapids processors by early 2022 because the company has at least one SPR-based supercomputer contract — the Crossroads — that is set to be delivered by the time.  

But if Intel launches Sapphire Rapids in early 2022, this leaves the company’s 3rd generation Xeon Scalable ‘Ice Lake-SP’ processor about a year of active life on the market. Meanwhile, the two platforms are completely incompatible and it will take Intel’s server partners quite an effort to transit to it. In addition to processor innovations, the new platform will support DDR5 memory, the PCIe 5.0 bus with a 32 GT/s data transfer rate that is enhanced with CXL 1.1 protocol to optimize CPU-to-device (for accelerators) as well as CPU-to-memory (for memory expansion and storage devices) interconnects.  

(Image credit: Intel)

Just like Alder Lake CPUs and Xe-HP GPUs, Sapphire Rapids will be made using Intel's 10nm Enhanced SuperFin process technology.

  • Geef
    Who wants to bet their Q4 timeframe will be 'pushed back' and they will tide everyone over with a 14++++++++++++++ chip?
    Reply
  • InvalidError
    Geef said:
    Who wants to bet their Q4 timeframe will be 'pushed back' and they will tide everyone over with a 14++++++++++++++ chip?
    I doubt Intel can afford any further delays when it looks like Zen 3 will pass Intel on per-core performance.
    Reply
  • JayNor
    Intel's Rocket Lake-S should take care of the zen3 desktop challenge.

    Intel also has pcie4 on their new laptop chips.
    I read an article for Samsung 980 Pro pcie4 m.2 format drive today. Looks like pcie4 is a good feature to have.

    https://www.phoronix.com/scan.php?page=article&item=samsung-980-pro&num=1
    Reply
  • Jimbojan
    Intel really should go directly to 5nm, that is to skip 7nm all together, if it can to catch up with and surpass TSMC
    Reply
  • InvalidError
    Jimbojan said:
    Intel really should go directly to 5nm, that is to skip 7nm all together, if it can to catch up with and surpass TSMC
    The "nm" figures are little more than marketing. Intel's 14nm process is about as dense as others' 10nm, its 10nm process is about as dense as others' 7-8nm and its 7nm process is expected to be about as dense as TSMC's 5nm.

    Also, even if Intel wanted to skip steps, it wouldn't be able to simply due to the limited supply of next-gen EUV equipment, so it would still need to deploy 7nm in the meantime just to keep up with demand that 10-14nm cannot keep up with.
    Reply
  • Kamen Rider Blade
    RocketLake is based off the same MicroArchitecture as TigerLake, and after seeing it go up against Renoir, I'm not impressed.

    AlderLake is where Intel has a real chance to catch up.
    Reply
  • watzupken
    So now Intel is going to call their 10nm+++ "enhanced SuperFin". I am not sure if they are going to regret going down this naming convention because 7nm is still a number of years away. So 10nm ++++ is going to be "very enhanced SuperFin". Whatever SuperFin they have, I don't feel its enough to turn the tide for them at this point. Especially so when they are unable to produce enough of 10nm chips.
    Reply
  • watzupken
    Kamen Rider Blade said:
    RocketLake is based off the same MicroArchitecture as TigerLake, and after seeing it go up against Renoir, I'm not impressed.

    AlderLake is where Intel has a real chance to catch up.
    Alder Lake may do well in the mobile space where the small cores will help. On the desktop space, its fairly pointless considering the rumored specs includes Alder Lake for desktop that does not come with any small cores at all. But if one is looking for performance, the lack of faster cores may limit their attractiveness. 8 fast and 8 slow cores for example will not bring them near the multicore performance of a proper 16 core Zen 3 CPU.
    Reply
  • JayNor
    watzupken said:
    8 fast and 8 slow cores for example will not bring them near the multicore performance of a proper 16 core Zen 3 CPU.
    -

    Intel's Gracemont doesn't have to match high performance cores ... just the AMD cores that have only avx2 and no dlboost. If they match AMD's zen3 cores doing SIMD and exceed AMD's zen3 doing simd AI operations with an atom core that is 1/3 the size and half the power of Intel's Golden Cove cores, then Intel can claim a win in many benchmarks, just as they demonstrated in the Tiger Lake launch.
    Reply
  • evilpaul
    Maybe they'll be really good in Sysmark, too? Are most people really running AI workloads and whatever software actually uses AVX512 on their laptops? The HPC laptop market?

    And PCIe 4.0 in laptops? Sounds amazing for battery life. Are there benchmarks of it making a difference outside of CrystalDiskMark' sequential QD32 tests?
    Reply
  • JayNor
    Sapphire Rapids status was stated as "broadly sampling" at sc20.

    It is on the 10ESF process ... same as the Xe-HP. It includes pcie5/CXL, DDR5, AMX tiled matrix processing of bfloat16, support for a new Optane version.

    Xe-HP is also sampling.
    Reply