Intel’s Alder Lake processors are coming soon with rumour-mills hinting at a November 4 launch. The Santa Clara, California-based chipmaker has been hyping the chip for a while now, particularly its ‘hybrid’ architecture. The idea is to offer a combination of performance (codenamed Golden Cove) and efficiency (Gracemont) x86 cores—instead of just next generation powerful CPU cores—for a mix of power when it’s needed and efficiency during less demanding tasks. Intel has previously teased a full array of chips from 9W to 125W that would make use of the new hybrid model which is to say that Alder Lake is much more ambitious than last year’s Lakefield chip.
As it should be, because competition is only getting stronger. Intel isn’t the only company to take the hybrid route for optimised computing. Arm’s big.LITTLE and DynamicIQ technologies have been around for over a decade. They serve as blueprints for Qualcomm’s Snapdragon chips that power the majority of Android smartphones in the market today and Apple’s A-series chips for iPhone. Perhaps even more relevant to this comparison, is Cupertino’s M1 chip for Mac that’s just received a big refresh with M1 Pro and M1 Max both of which feature eight high-performance and too high-efficiency cores.
| ‘One size can’t fit all’: Intel’s Prakash Mallya on why the chipmaker is making discrete GPUs now
Needless to say, a lot will be riding on how Intel’s Alder Lake chips turn out when they start shipping this year—desktop first, followed by mobile. TheSpuzz Online caught up with Intel’s Prakash Mallya to understand some of the nitty-gritties of Alder Lake, its significance for the company in times of increasing competition and the broader state of affairs of the semiconductor industry, aka the global chip shortage and when will it end. Excerpts.
FE | Take us through Alder Lake, its feature set and improvements over last gen.
Prakash Mallya | Look, it’s a fantastic product. It is a mix of Performance and Efficient cores. The Efficient core is a highly scalable x86 microarchitecture for addressing compute requirements across the entire spectrum of our customers’ needs, from low power mobile applications to many-core microservices. It was designed for throughput, enabling scalable multi-threaded performance for modern multi-tasking. Compared to Skylake, the Efficient core delivers 40% more performance at the same power, or the same performance while consuming less than 40% of the power.
When you look at the Performance core, this x86 core is not only the highest performing CPU core Intel has ever built, but it also delivers a step function in CPU architecture performance that will drive the next decade of compute. It was designed for speed, low latency and single-threaded application performance.
This high-performance hybrid approach is further enhanced by Intel Thread Director, a technology designed into Alder Lake that helps the operating system make more intelligent and informed decisions about where to place running threads. It uses real-world hardware telemetry to guide operating systems to the right workload, which can be guided to the right cores at the right time. Thread Director is fully dynamic, adaptive, and autonomous versus a static, deterministic, software-only approach.
FE | How does your hybrid model stack up against competition and what are some of its advantages?
Prakash Mallya | Our design approach to Alder Lake starts with delivering incredible performance that scales from ultra-portable laptops to enthusiast and commercial desktops. We are introducing two completely new microarchitectures for Alder Lake and both are high performance on their own and complement each other. The conventional approach of ‘hybrid’ currently in the industry is often focused on getting the most hours out of a battery which often requires trading off performance for power efficiency.
A good analogy is to think of conventional hybrid cars that focused on getting the most miles out of a tank of gas versus Formula 1 racing cars that use hybrid technology to achieve maximum performance—they have engines to give them top speed and then electric power to blast them out of the corners with acceleration. Alder Lake is the high-performance hybrid that we built for computing.
It will be Intel’s first processor built on the Intel 7 process and will serve as the foundation for leadership desktop and mobile processors that deliver smarter, faster, and more efficient real-world computing in a variety of form factors. It will begin shipping this year and we’ve said desktop will be followed by mobile. We’ll have more specifics to share later.
I will say though, when people said PCs are a mature category, in my opinion, PC innovation is at its highest that I’ve ever seen. I’ve been 21 years in the company and I’m constantly in awe of the amount of innovation being brought together on PC and Alder Lake is a great example of that.
FE | What will be its use case? Is there any category that will benefit ‘specifically’ from these new chips?
Prakash Mallya | Because of the hybrid nature of the platform, it caters to a variety of needs and all those needs are critical. You can say one need is commercial applications. If you’re building an enterprise notebook or enterprise desktop, you require great performance, very good battery life, great screen quality, and flexible design. Can Alder Lake do that? Yes, it can. Similarly, for consumer or commercial desktops, it can be game-changing, because the performance threshold has a material impact on all application performance. I can’t say one segment will be better served than the other. Alder Lake will alter the way people look at PC computing in a big way.
FE | What is Thread Director’s role in all this?
Prakash Mallya | One of the biggest reasons of Intel’s success in the 52 years is the level of software innovation that we do along with the ecosystem. We don’t do this alone. The technology that we built on Centrino, which was wireless, a lot of it was power management and we brought the elements of that together along with the industry. Ultrabooks was the evolution that we led a few years back, that again had different elements, one of them was software and then it had elements of design, form factor, and screen. When you look at even today, the enabling that we do on software across every platform—not only Alder Lake—people have seamless experience on x86 Intel-based platforms for that reason. Now, I will tell you specifically on Thread Director, given it’s a hybrid architecture, and that it is dynamic in nature, the flexibility of it being able to run multi-threaded, to single-threaded in a dynamic basis, because of telemetry that happens dynamically, it is very powerful and the kind of work we will do with the software ecosystem and continue in the future will have a material impact on the experience that we enable for consumers or business owners or gamers or otherwise.
FE | How crucial is software to enhancing the capabilities of the silicon?
Prakash Mallya | OpenVINO as an edge AI inference track is a great example of what software can do to really open up the features and drive value for the people building edge inference-based applications. It is a software toolkit. The kind of work that we are doing on network, whether it is the software stack that we have built and reference platforms that we built, which makes it possible for either the telecom service providers or the solutions providers in the telecom space to bring alive the feature sets that are built into the silicon, whether it is the N SKUs, which are the Xeon SKUs, which combined with the software, really accelerate the network workload performance. The software work happens in every part of our architecture and with Intel Innovation, what our CEO Pat Gelsinger is signaling to the industry is that software is going to be extremely important. Software is going to be very important and material to the strategy of the company.
FE | What is Intel’s stand on the ongoing global semiconductor shortage—how are you placed, what are you doing to ramp up manufacturing and when will this crisis end?
Prakash Mallya | Technology has never been more important than it is today. Lockdowns and work-from-home policies during the COVID-19 pandemic greatly increased the need for laptops, gaming consoles, IoT devices and more. This caused an explosive growth in semiconductors that has placed great strain on the global supply chain. It will take a few years to address shortages of foundry capacities, substrates and components.
We are actively working with our supply chain partners to increase the availability of third-party materials and components to further improve output for our processors and support the broader PC ecosystem. We’ve been expanding our capacity to meet demand and, as a result, we expect to grow our annual client CPU supply by double-digits year-over-year versus 2020.
We are expanding our advanced chip manufacturing capacity by investing $20 billion in two factories in Arizona. We plan to expand our presence in the U.S. and in Europe to ensure a sustainable and secure semiconductor supply chain for the world. We remain focused on supporting our customers and we will continue working to increase supply to meet their needs.
| Rocket Lake-S deep dive: Intel on broken embargoes, benchmark scores, walled gardens and power of PC