A theme of the new AMD is modularity. We've of course heard this before as it has always been a goal of AMD's to bring to market more modular, configurable designs, however this time the rhetoric is a lot more serious. In our earlier coverage we talked about future AMD SoCs allowing for a combination of AMD x86 CPU, GPU and 3rd party IP blocks. What AMD didn't mention during its Financial Analyst Day presentations however was how it would enable this pick-and-choose modular design. The secret, as it turns out, is in a new modular fabric that AMD is designing. 

It will take AMD until 2014 - 2015 to actually have the first, fully functional modular fabric in an SoC, but that's the goal. Being able to design a foundation that can interface with multiple buses (e.g. PCIe, HT, AMBA for ARM, etc...) will enable AMD to build more modular SoCs. 

With the fabric created, AMD can also change the way it does chip design. Today APU designs are seen from start to finish. Teams work on the various components of the design, but those components are viewed as a part of the whole, not as independents. E.g. the GPU portion of Trinity is worked on as Trinity's GPU, not a GPU block that will be re-used in other chips. Under the new AMD, teams will work on designing modular IP blocks without much focus on where they end up. You'll have teams that will work on a GPU block and simply move onto another GPU project after they're done.

Assuming AMD's new scalable SoC fabric is flexible enough, theoretically an APU designer could pick and choose from the various IP blocks and deliver a customized design that's decoupled from the individual blocks themselves. Similar to how you'll see an Imagination Technologies PowerVR SGX 540 in a variety of SoCs, AMD could build a GCN GPU block and use it in a variety of SoCs that address different markets. You can view AMD as having a broad portfolio of x86 and GPU cores and with this new SoC fabric it can mix and match those blocks as it sees fit. Furthermore, if the need arises, AMD could add in 3rd party IP where appropriate. 

We've actually heard of similar approaches to design from other companies in the SoC space, including Intel. With Atom Intel introduced a sea-of-FUBs (functional unit blocks) design methodology that leveraged more synthesized logic and modular blocks to reduce time to market and reduce feature creep. Atom also uses a fair amount of 3rd party IP (GPU, video encode/decode).

AMD's strategy makes a lot of sense. There's still a lot of execution that needs to happen before we get to the point where we can take modularity for granted, but the direction is sound.

POST A COMMENT

18 Comments

View All Comments

  • chizow - Friday, February 03, 2012 - link

    Idk it just seems awfully short-sighted to ignore that market since that's where all the gadget growth is projected over the next 3-5 years. Tablets and smartphones.

    Intel's Atom started off modestly, shooting for mini-ITX HTPCs, same as Nvidia's Tegra as a drop-in mGPU with Atom. Keep in mind, Intel also resisted the idea of Netbooks but at least they had a contingency plan with Atom. Nvidia obviously wanted to get away from the idea of x86 as quickly as possible and embraced ARM.

    It looks like AMD is also getting interested in at least the Tablet market, but the roadmap indicates they won't even have something that fits that TDP profile until next year. Just way behind even getting their foot in the door, but I think Reed will at least have them in a better position to compete through more timely execution.
    Reply
  • cfaalm - Friday, February 03, 2012 - link

    Some of the reactions here seem to assume all at AMD are gonna think. "OK we did the slide show, now let's figure out how to do all this." You bet there's stuff in the works already. You can't come up with these slides based on nothing but ideas. I just hope like many of you that their execution will be timely from now on. Reply
  • fujiyama - Saturday, February 04, 2012 - link

    For last 25 years Intel was successful to deliver better and better CPUs with AMD following.
    Introducing Athlon CPU AMD they took lead and profits limited only by FAB and Intel bribes.
    Mixing GPUs, CPUs etc. says about INABILITY to create a competitive core.
    Adding GCN, ARM and I/O doesn't change that.
    Reply
  • silverblue - Saturday, February 04, 2012 - link

    Much like Intel slapping a GPU on their CPUs doesn't make their GPUs any good. Reply
  • Sunsmasher - Saturday, February 04, 2012 - link

    These slides are very disturbing in that they have tons of "Corporate Speak" platitudes that don't indicate that AMD has any real ideas for the future.
    This is just corporate boardroom "noise" devoid of any real content.
    Unfortunately, this furthers the suspicion that AMD is really flailing and has no real
    Idea of where to go as a company.
    Reply
  • haplo602 - Sunday, February 05, 2012 - link

    Didn't they do this in the past already with Torrenza ? All it achieved were a few very specific products. I don't see this getting past initial designs ... Reply
  • andreamaeB - Monday, February 06, 2012 - link

    I like it. Just a few more innovations and it was perfect. Reply
  • hingfingg - Thursday, February 16, 2012 - link

    ** {{w w w }} {{proxy4biz }} {{ com}} ***** Reply

Log in

Don't have an account? Sign up now