It has been hard to miss the fact that Intel has been vacuuming up a lot of industry talent, which brings with them a lot of experience. Renduchintala, Koduri, Keller, Hook, and Carvill, are just to name a few. This new crew has decided to break Intel out of its shell for the first time in a while, holding the first in a new tradition of Intel Architecture Days. Through the five hours of presentations, Intel lifted the lid on the CPU core roadmaps through 2021, the next generation of integrated graphics, the future of Intel’s graphics business, new chips built on 3D packaging technologies, and even parts of the microarchitecture for the 2019 consumer processors. In other words, it's many of the things we've been missing out on for years. And now that Intel is once again holding these kinds of disclosures, there’s a lot to dig in to.

Contents List

Intel covered a good amount of ground at the Architecture Day, which we’ve split into the following categories:

  1. The CPU Core and Atom Roadmaps, on 10nm
  2. The Sunny Cove Microarchitecture
  3. The Next Generation Gen11 Graphics
  4. Intel Demonstrates Sunny Cove and Gen11 Graphics
  5. Beyond Gen11 Graphics: Announcing the Xe Graphics Brand
  6. 3D Packaging with FOVEROS
  7. Intel’s first Fovoros and first Hybrid x86 CPU: Core plus Atom in 7 W on 10nm
  8. Ice Lake 10nm Xeon
  9. Intel Made Something Really Funny: Q&A with Raja, Jim, and Murthy

 

The CPU Core Roadmaps

It is common for companies like Intel to ask members of the press what they enjoy about announcements from Intel, Intel’s competitors, or other companies in the industry. One of answers I will never tire of saying is ‘roadmaps’. The roadmap is a simple document but it enables a company to explain part of its future plans in a very easy to understand way. It shows to the press, to customers, and to partners, that the company has a vision beyond the next product and that it expects to deliver at a rough cadence, hopefully with some markers on expected performance additions or improvements. Roadmaps are rarely taken as set in stone either, with most people understanding that they have an element of fuzziness depending on external factors.

To that end, I’ve been requesting Intel to show roadmaps for years. They used to be common place, but ever since Skylake, it has kind of dried up. In recent months Intel has shown rough datacentre roadmaps, with Cascade Lake, Cooper Lake, and Ice Lake and the next few generations. But for the Core family it has been somewhat more difficult. Depending on which analyst you talk to, a good number will point to some of the Skylake derivatives as being holding points while the issues with 10nm have been sorted out. But nonetheless, all we tend to hear about is the faint whisper of a codename potentially, which doesn’t mean much.

So imagine my delight when we get not one roadmap from Intel on CPUs, but two. Intel gave us both the Core architecture roadmap and the Atom architecture roadmap for the next few generations.

For the high performance Core architecture, Intel lists three new codenames over the next three years. To be very clear here, these are the codenames for the individual core microarchitecture, not the chip, which is an important departure from how Intel has previously done things.

Sunny Cove, built on 10nm, will come to market in 2019 and offer increased single-threaded performance, new instructions, and ‘improved scalability’. Intel went into more detail about the Sunny Cove microarchitecture, which is in the next part of this article. To avoid doubt, Sunny Cove will have AVX-512. We believe that these cores, when paired with Gen11 graphics, will be called Ice Lake.

Willow Cove looks like it will be a 2020 core design, most likely also on 10nm. Intel lists the highlights here as a cache redesign (which might mean L1/L2 adjustments), new transistor optimizations (manufacturing based), and additional security features, likely referring to further enhancements from new classes of side-channel attacks.

Golden Cove rounds out the trio, and is firmly in that 2021 segment in the graph. Process node here is a question mark, but we’re likely to see it on 10nm and or 7nm. Golden Cove is where Intel adds another slice of the serious pie onto its plate, with an increase in single threaded performance, a focus on AI performance, and potential networking and AI additions to the core design. Security features also look like they get a boost.

Intel Core Microarchitecture Roadmap
Core Name Year Process Node Improvements
Skylake 2015 14 nm Single Threaded Performance
Lower Power
Other Optimizations
Kaby Lake 2016 14 nm+ Frequency
Coffee Lake 2017 14 nm++ Frequency
Coffee Refresh 2018 14 nm++ Frequency
Sunny Cove 2019 10 nm Single Threaded Performance
New Instructions
Improved Scalability
Willow Cove 2020 ? 10 nm ? Cache Redesign
New Transistor Optimization
Security Features
Golden Cove 2021 ? 7 / 10 nm ? Single Threaded Performance
AI Performance
Networking / 5G Performance
Security Features

The lower-powered Atom microarchitecture roadmap is on a slower cadence than the Core microarchitecture, which is not surprising given its history. Seeing as how Atom has to fit into a range of devices, we’re expecting there to be a wide range in capabilities, especially from the SoC side.

The upcoming microarchitecture for 2019 is called Tremont, which focuses on single threaded performance increases, battery life increases, and network server performance. Based on some of the designs later in this article, we think that this will be a 10nm design.

Following Tremont will be Gracemont, which Intel lists as a 2021 product. As Atom is designed to continually push both the performance at the high-end of its capabilities and the efficiency at the low-end, Intel lists that Gracemont will have additional single threaded performance and a focus on increased frequency. This will be combined with additional vector performance, which likely means that Atom will get some wider vector units or support new vector instructions.

Beyond this will be a future ‘mont’ core (and not month as listed in the image). Here Intel is spitballing what this new 2023 core might have, for which the general listing of performance, frequency and features is there.

Intel Atom Microarchitecture Roadmap
  Year Process Improvements
Goldmont 2016 14 nm Higher Performance
Cryptography Features
Goldmont Plus 2017 14 nm Branch Prediction
More Execution
Larger Load/Store Buffers
More Cache
- 2018 - -
Tremont 2019 10 nm ? Single Threaded Performance
Network Server Performance
Battery Life
- 2020 - -
Gracemont 2021 10 nm ? Single Threaded Performance
Frequency
Vector Performance
- 2022 - -
'Next Mont' 2023 ? Single Threaded Performance
Frequency
'Features'

As stated above, these are just the microarchitecture names. The actual chips these cores are in will likely have different names, which means a Lake name for the Core microarchitecture. At the event, Intel stated that Ice Lake would have Sunny Cove cores in it, for example.

Another aspect to Intel’s presentations was that future microarchitectures are likely to be uncoupled from any process technologies. In order to build some resiliency into the company’s product line moving forward, both Raja Koduri and Dr. Murthy Renduchintala explained that future microarchitectures will not be process dependent, and the latest products will come to market on the best process technologies available at the time. As a result we’re likely to see some of the Core designs straddle different manufacturing technologies.

Intel also went into a bit of detail on microarchitecture of Sunny Cove.

Sunny Cove Microarchitecture: A Peek At the Back End
POST A COMMENT

149 Comments

View All Comments

  • zodiacfml - Thursday, December 13, 2018 - link

    YES Reply
  • Raqia - Thursday, December 13, 2018 - link

    For ultra-mobile, not only are battery/power/heat issues but supply is one as well due to Intel being locked down to their own manufacturing division. On top of that, they have a lock on x86 by not licensing to any competitors but AMD, who despite competitive stretches inevitably stumbles (either due to themselves or Intel's non-engineering financial efforts) and leaves the industry with dry spells of performance improvements. Intel's gross margins on their chips remain >60% as a result whereas ARM SoCs even after licensing is closer to 20-30%. Reply
  • Raqia - Thursday, December 13, 2018 - link

    Keller declared that the technology is in its infancy, and feature wise the 2019 version of the Atom simply won't be competitive with leading ARM SoCs like the 8cx. The slowness you refer to only occurs when running native 32 bit x86 code on the WOW emulation layer, but the value of this feature is mostly in the compatibility being there at all. If performance and compatibility of legacy code matters to you then certainly Windows on Arm isn't suitable. However, it will matter even less now with the new native compilation tools and ports of important sub platforms like Chromium. Reply
  • 29a - Thursday, December 13, 2018 - link

    "Windows on ARM is horribly slow and therefore shitty."

    Sounds a lot like Windows on Atom.
    Reply
  • MonkeyPaw - Saturday, December 15, 2018 - link

    I’m betting Apple wanted one for MacBook Air, or maybe MS for Surface Go. It would be the right amount of performance for both devices, an both companies would have the clout to get it done. I’d lean toward Apple because the GPU is pretty big. Reply
  • Kevin G - Wednesday, December 12, 2018 - link

    Typo:
    "a physical address space up to 52 bits. This means, according to Intel, that the server processors could theoretically support 4 TB of memory per socket."

    That should be petabytes instead of terabytes. The limit is for an entire system, not per socket as additional sockets will not grant any additional capacity.
    Reply
  • gamerk2 - Thursday, December 13, 2018 - link

    NUMA systems could potentially be per-socket rather then OS wide. Reply
  • HStewart - Wednesday, December 12, 2018 - link

    It sounds like Intel has been working on increasing performance in two ways
    1. 7nm change for the future - because of limitations found with 10nm
    2. 10nm enhance for corrections for performance of issues with Cannon Lake

    But most importantly, architexture improvements like faster single thread execution and new instructions and multi-core improvements will in long term significantly improve performance
    Reply
  • ishould - Wednesday, December 12, 2018 - link

    Forgive me if I take 2 metric tons of salt with any roadmaps Intel provides these days. They haven't exactly had the most accurate timelines as of late (past four years) Reply
  • HStewart - Wednesday, December 12, 2018 - link

    It appears they realize that and coming out with document to indicated they have made corrections - this is better than not knowing what they are planning - or as some AMD Fans would like to believe that they lost the battle. Reply

Log in

Don't have an account? Sign up now