Three weeks after Intel (INTC) – Get report shares tossed on news of a decline in manufacturing technology, the company shares lots of details on how it aims to compete with AMD (AMD) – Get report and Nvidia (NVDA) – Get report in the coming years.
At this year’s edition of the annual Architecture Day event, Intel offered new details about the soon-to-be-launched Tiger Lake notebook processor, as well as its next-gen Alder Lake PC CPU platform and its plans to battle AMD and Nvidia in hosting GPU markets .
And most of all ̵
Intel claims that SuperFin, which will be used by Tiger Lake processors and Intel’s first discrete GPUs, will deliver 17% to 18% better transistor performance than Intel’s first 10nm transistors delivered.
The company also said it works with “Enhanced SuperFin” transistors; they will deliver additional performance benefits and will be used by both future GPUs and Intel’s Sapphire Rapids server CPU platform, due in the second half of 2021. But none of the versions of SuperFin will be used by Intel’s Ice Lake server CPU line, which will begin shipping by the end of the year and will compete with AMD’s expected third-generation Epyc server CPU line (Milan).
Technical analyst Patrick Moorhead noted that Intel’s latest advances seem to give it an industry leader in transistor performance. Taiwan Semiconductor (TSM) – Get report, which recently started mass production for its 5 nm node (seen as competitive with Intel’s planned 7 nm node), currently has a clear lead in transistor density.
Tiger Lake and Alder Lake
Thanks in part to “dramatic” clock speed improvements enabled by SuperFin, Intel promises that its 10nm Tiger Lake line, which will be unveiled at a September 2 event and rely on a new CPU core microarchitecture known as Willow Cove, will deliver. ” more than a generation increase in CPU performance. ”The company also reiterates that Tiger Lake’s integrated GPUs – based on a new architecture known as the Xe-LP – will deliver a 2x performance improvement over Intel’s older Gen11 integrated GPUs: you, and says Tiger Lake will provide “massive” performance gains for processing AI / deep learning algorithms.
If Intel succeeds in its performance / clock speed requirements, Tiger Lake should put Intel on a stronger footing in the portable processor market, with AMD seeing strong growth following the launch of its Ryzen Mobile 4000 line in January. Last month, AMD said revenue from notebook processors more than doubled annually during the second quarter.
Regarding the 10nm Alder Lake line, which will take place in the second half of 2021 and is intended for both desktops and laptops, Intel said that the CPUs (like many ARM architecture CPUs) will include a blend of high-performance cores based on a near-through microarchitecture known as Golden Cove and low-power cores based on a microarchitecture known as Gracemont.
Such an architecture can help improve the battery life of laptops. However, there is some skepticism about the benefits of using it in the high-end desktop CPU market, where AMD has become quite competitive and performance tends to mean more than power consumption. Until Alder Lake arrives, Intel will continue to rely on its older 14nm desktop node process node – the company is believed to be working on a new 14nm desktop CPU line, known as Rocket Lake.
Separately, Intel removed wraps from Client 2.0, a long-term initiative to develop PC processors whose various processing and I / O functions are placed on several different chips and integrated using Intel’s packaging and interconnection technologies. The company claims that Client 2.0 can reduce processor development times by a year or so compared to creating a traditional, monolithic processor, while allowing Intel to mix and match manufacturing processes and reuse IP chips.
Intel GPU plans
For the discrete GPU market, Intel promises to launch more offerings between now and the end of 2021. Specifically:
- Xe-HP and Xe-HPC, a pair of high-performance server GPUs coming next year. With the exception of an Xe-HP variant, GPUs will rely on multiple “chips” connected via high-speed connections. Xe-HPC, codenamed Ponte Vecchio and thought to be Intel’s flagship server GPU, will rely on a mix of Intel and third-party manufacturers.
- Xe-HPG, a gaming GPU architecture that will support real-time radiation tracking and promises to meet the needs of “enthusiast gamers.” It will be manufactured by a third party (possibly TSMC) and will come sometime next year.
- DG1, a discreet PC GPU for developers that first appeared in January and is promised to be delivered by the end of the year.
- SG1, a server GPU that stitches together four DG1 GPUs and is expected to be delivered this year. Intel says SG1 is intended to support “low latency, high density Android cloud games and video streaming.”
GPU announcements come three months after Nvidia, which remains dominant in the server GPU market, unveiled a versatile flagship server GPU, known as the A100, which relies on its new Ampere architecture. And that comes just before a September event where Nvidia is expected to unveil its first Ampere gaming GPUs.
AMD is currently preparing game GPUs based on a new architecture known as RDNA 2, as well as a server GPU based on a new architecture known as CDNA. Both expire at the end of the year.
Intel’s share has fallen by 1.5% in Thursday’s trading, amid broader weakness in companies with exposure to hardware for companies after Cisco Systems’ (CSCO) – Get report soft October quarterly guidance.