AMD engineers have apparently spent a lot of time discussing the big.LITTLE architecture from Arm. Although Intel is looking to implement it across all its client lineups starting with the 12th Gen Alder Lake-S CPUs, AMD won’t be considering it for the time being.
The reason being that it’s simply not worth it, at least not for the next 4-5 years. As per Joe Macri, CVP, and Product CTO at AMD, Team Red won’t increase the core counts just for the marketing benefits and will stick to a uniform core design for the next few architectures.
We’ve been studying big.LITTLE. It’s been 15+ years, so this is not a new concept in any way shape or form. We continue to study it, we continue to look at it. We’re not going to talk about whether we’ll do it or not, but I’m going to talk about some of the challenges of it and around what you really want to do with it. Is the goal power efficiency? Is the goal more performance? Is the goal just marketing, ‘I want more core count’, regardless of what it does for the other two variables? At AMD, the marketing one we’re going to throw out the window. We’re not going to do it just to have a bigger number.Joe Macri, CVP and product CTO
That certainly is a shot at Intel’s strategy, or perhaps the lack thereof. Seeing how Lakefield performed on Windows 10, AMD certainly has a point, but Alder Lake and the hybrid core architecture is still at least 2 years away. Will Intel be able to convince developers, especially game studios to optimize their titles for their new CPU design by then? Probably not, as the consoles run on AMD Zen 2 CPUs, and adding support for a heterogeneous core architecture will require quite a bit of work at the engine level.
Just driving up the core count with little isn’t going to be that useful until software comes along. It’s not a simple problem, and I think there’s been at least one company who has got it right. I wouldn’t say all the companies have got it right that have started doing it. What we’re really fighting, for big.LITTLE with AMD, is that Mike’s team is doing such a great job with big, it makes it hard to win with LITTLE.Joe Macri, CVP and product CTO
Furthermore, seeing how things are, if Intel continues to lose its market share in the consumer segment as it is, it’ll be on par with AMD by the time Alder Lake is launched. If the latter is able to hold onto its lead in gaming or even level with Intel, then devs will see no point in optimizing their game engines for the latter.
Over time I think there will be a point when we are going to need LITTLE, and it will be a point in time when the OS has the right attributes, the right capabilities in its scheduler and memory allocator, we’ll have the right memory subsystem. We’ll be able to give you not just a little bit better experience, but a much better experience. If we can’t get that experience change to be noticeable, why do it?Joe Macri, CVP and product CTO
Content creation and lightly threaded workloads such as browser scripts, word processing, and media encoding will likely get optimized for heterogeneous architectures as the chips are rolled.
Considering all that, it’s very likely that AMD will transition to the big.LITTLE by the time Intel is done optimizing the x86 Eco-system for the hybrid core design. All in due time.