All your base station are belong to us: Intel joins spec race with new 5G chips

Plus: Looking for a 'deep learning boost'? Chipzilla's your man


Intel will sling out fresh silicon aimed at hardware across the nascent 5G market: including new second-generation Xeon processors and a low-latency Atom P5900 for next-gen mobile base stations.

The semiconductor giant, which had been planning to launch its new chips at the now-cancelled Mobile World Congress in Barcelona, also flagged up its first all-new structured application-specific integrated circuit (ASIC) for 5G, codenamed Diamond Mesa, and a new network adapter that is also tailored for the high-bandwidth, low-latency needs of 5G tech.

Intel claims the updated second-generation Xeon Scalable processor, now at a top speed of 3.9GHz, will give a 36 per cent boost in performance and a jump of 42 per cent in performance per dollar spent versus its first generation silicon (translation: the new chips will be expensive). It also said the processor will be the first it has created that "has AI built in" giving a "deep learning boost" – claiming that the DL processing specifically takes place "up to six times faster" than its rivals.

Someone using a laptop on a couch

Intel's back. Can't keep it down. Back with 5G. Back in the game, back with modems... that have 'MediaTek' written on them

READ MORE

"No one is better placed for 5G success," said one Intel exec on a pre-briefing of the news, saying that the server chip was "tailored to network workloads".

As for the 10nm process base-station chip, the P5900, it is also designed for high-bandwidth, low-latency work and Intel claimed it is 3.7 times more efficient at packet balancing throughput. Intel said it will start appearing in new base stations from Ericsson, Nokia and ZTE later this year, and Chipzilla predicted it will be in six million base stations by 2024.

The company didn't provide release dates or prices but representatives did set a world record for buzzword bingo, talking about "cloudification" and "industry inflections" while opining about "multi-function compute hungry devices" and how the new chips would "unleash the full potential of data".

Despite the years of hype around 5G and the "race to 5G", the reality is that the infrastructure for the new standard is only now starting to be rolled out. Current 5G deployments are limited to cities and are non-standalone, meaning they piggyback on 4G kit; moreover, there is not a great deal of 5G-capable hardware using the tech. You can expect vendor announcements that were meant for the cancelled Mobile World Congress to change that soon.

Telco standards bods at 3GPP are currently working on studies into improving the upcoming standalone standard's efficiency – with Phase 2 (release 16) of the 5G system at its Freeze date next month, and due to reach completion in June 2020.

Intel quit making 5G smartphone modems back in April '19 after its hopes of getting its cellular modems into iPhones were dashed because Qualcomm and Apple inked a peace pact on their patent and licensing battles across the world. Intel went on to sell its 5G modem biz to Cupertino shortly after. But silicon for the base stations, servers and switches providing the backend of 5G is still very much up for grabs.

Intel has dominated the server chip world until recently, with supply constraints and fab line upgrades hitting availability. Its introduction of the new Gen 2 scalables comes against the backdrop of an ongoing industry-wide, 18-month-plus shortage of some of its Cascade Lake parts that is predicted to continue all the way through 2020. ®

Similar topics


Other stories you might like

  • Intel freezes hiring for PC chip team, cites 'macroeconomic uncertainty'
    Inflation, Apple M2, PC market shrink: Could the timing have been worse?

    Intel's PC chip division is the latest team caught in the current tide of economic uncertainty, as the company freezes hiring in the group. 

    In an internal memo obtained by Reuters, Intel told employees all hiring and job requisitions in the client computing group were on hold for at least two weeks. During that time, the chipmaker will reportedly be reevaluating its priorities with "increased focus and prioritization in our spending [to] help us weather macroeconomic uncertainty," Intel said. 

    The client computing group, which designs end-user hardware, is Intel's largest by sales, having generated $9.3 billion of the $18.4 billion Intel made last quarter. Despite its place at the top, the CCG's Q1 takings were still down 13 percent compared to the same time in 2021. It was also the only Intel division to lose money compared to Q1 2021, another potential reason for the hiring freeze in the sector. 

    Continue reading
  • Apple gets lawsuit over Meltdown and Spectre dismissed
    Judge finds security is not a central feature of iDevices

    A California District Court judge has dismissed a proposed class action complaint against Apple for allegedly selling iPhones and iPads containing Arm-based chips with known flaws.

    The lawsuit was initially filed on January 8, 2018, six days after The Register revealed the Intel CPU architecture vulnerabilities that would later come to be known as Meltdown and Spectre and would affect Arm and AMD chips, among others, to varying degrees.

    Amended in June, 2018 the complaint [PDF] charges that the Arm-based Apple processors in Cupertino's devices at the time suffered from a design defect that exposed sensitive data and that customers "paid more for their iDevices than they were worth because Apple knowingly omitted the defect."

    Continue reading
  • Intel to get $7.3b for Germany fab site as TSMC dismisses Europe plans
    x86 giant giddy about making chips on the continent, its foundry rival not so much

    Intel is reportedly set to receive €6.8 billion ($7.3 billion) in subsidies for a massive chip manufacturing campus it's planning in Germany, and the x86 giant apparently won't have to worry about foundry rival TSMC setting up shop anywhere nearby for the time being.

    The German subsidies for Intel's planned fab site in Magdeburg was disclosed last week by Martin Kröber, the city's representative in the Bundestag, according to local media. The federal government has already allocated €2.7 billion in its 2022 budget [PDF] for the project, according to Kröber.

    Germany's Deutsche Presse-Agentur said the government is discussing the possibility of subsidies for other projects in the microelectronics industry.

    Continue reading
  • Nvidia taps Intel’s Sapphire Rapids CPU for Hopper-powered DGX H100
    A win against AMD as a much bigger war over AI compute plays out

    Nvidia has chosen Intel's next-generation Xeon Scalable processor, known as Sapphire Rapids, to go inside its upcoming DGX H100 AI system to showcase its flagship H100 GPU.

    Jensen Huang, co-founder and CEO of Nvidia, confirmed the CPU choice during a fireside chat Tuesday at the BofA Securities 2022 Global Technology Conference. Nvidia positions the DGX family as the premier vehicle for its datacenter GPUs, pre-loading the machines with its software and optimizing them to provide the fastest AI performance as individual systems or in large supercomputer clusters.

    Huang's confirmation answers a question we and other observers have had about which next-generation x86 server CPU the new DGX system would use since it was announced in March.

    Continue reading

Biting the hand that feeds IT © 1998–2022