The AMD Embedded G-Series platform being introduced tonight is the world's first Accelerated Processing Unit (APU) for embedded systems. AMD has had quite a bit of history of supporting x86 based embedded systems. Starting with the Geode processor in 2003 (obtained from National Semiconductors and used in the OLPC project), AMD went on to introduce AMD64 technology into the embedded markets with the AMD Opteron processors in 2005. In 2007, the addition of graphics and other chipset options by AMD enabled comprehensive embedded solutions. In 2009, AMD introduced BGA (Ball Grid Array) packaging to meet customer demand.

At CES 2011, they gave us a sneak peek into the Embedded G-Series platform based on Brazos. AMD has increased performance and features in every generation while bringing down the power, area and price barriers for x86 in the embedded market.

The embedded market space is dominated by SoCs based on RISC processors such as ARM and MIPS. For most power sensitive embedded applications, PowerPC and x86 based solutions do not make the cut. x86, in particular, has been the dark horse due to the excessive power consumption for systems based on that architecture. Process shrinks have helped lower the power consumption numbers. However, we are still a few nodes away from when the x86 based solutions can really compete with RISC based solutions on the power front.

In the meantime, solutions like what we are seeing from AMD today integrate premium graphics capabilities within power envelops similar to what x86 used to consume in the previous generation—so you get CPU+GPU instead of just a CPU. RISC based embedded solutions may still be winning on the power front; however, for applications where slightly higher power consumption is not a concern, the x86 threat from the AMD embedded G-Series platform can become a cause for concern. MIPS is usually popular in such applications (set top boxes, digital signage etc.) and they will be facing credible opposition with AMD's integrated graphics capabilities.

The AMD Embedded G-Series
Comments Locked


View All Comments

  • semo - Wednesday, January 19, 2011 - link

    I think that Intel use the term to describe CPUs with extended support too. Yep, lost all meaning as far as I'm concerned
  • int9 - Wednesday, January 19, 2011 - link

    The T40N and the T44R are 9W but substantially weaker. This seems like another blow against Atom. I can't wait to see what that GPU will do.
  • silverblue - Wednesday, January 19, 2011 - link

    "This new "class" of CPU's are nothing more than downscaled desktop counterparts."

    But... it's the same as the desktop equivalent.

    I agree with the idea that it's hungrier than it should be for this sort of thing, however please bear in mind that it's not going to be drawing 18W constantly. Power gating and downclocking should ensure a far lower average usage.
  • ninjaquick - Wednesday, January 19, 2011 - link

    Embedded is a form-factor matter. An embedded solution can also be a slave system. An example of embedded use for this would be automotive interface computers, like built in GPS or Media Players. Embedded basically means it is not used as a traditional computer.
  • driscoll42 - Wednesday, January 19, 2011 - link

    After reading the comments I wanted to just say that there are some of us who did appreciate the article. I understand that you don't have a whole lot to go off of other than the Press Release so expecting many more details is a bit far fetched. I prefer some context to a Press Release as opposed to just the PR itself. As a student just starting a Computer Architecture course it's always exciting to what's new in the world, especially in embedded since that's the focus of the course. Anyways, thanks for the article, I look forward to reading more.
  • digitalzombie - Wednesday, January 19, 2011 - link

    Article seems to lean toward AMD because of all the AMD's supplied pictures and you're talking about the potential of AMD's embedded product. I would like to point out that you did state negatives for AMD's in the embedded market. Just fyi on why people might think this is an AMD's PR. There is probably nothing you can do, other than maybe photoshop some funny cute cat pictures with AMD on it.

    How come AMD and Intel are so behind in the embedded market? Is it because of their lack of interest until now and their non-existing technologies for low power chip? Their lack of IP's in technologies such as FPGA or whatever fancy do hicky? Is it because building cpu for the embedded systems require a different set of skills that AMD and Intel aren't well verse as ARM and MIPS? I guess, I don't understand how these major CPU company AMD and Intel cannot compete or beat ARM and MIPS. All they do is make CPU, well they're doing GPU now I guess, how can they not beat these smaller companies? Is the x86 ISA are too complex for embedded system? Doesn't make sense, I guess, since MIPS have 64bit.
  • silverblue - Wednesday, January 19, 2011 - link

    That's almost like asking "why can't ARM and MIPs compete with Intel and AMD in the desktop space?". I don't know... that's just the way I see it.

    If these are the same as the other Fusion processors, they'll fully support the x86-64 architecture. Oddly enough, they also support AMD-V... :)
  • Sufo - Wednesday, January 19, 2011 - link

    "why can't ARM and MIPs compete with Intel and AMD in the desktop space?"

    They can, and will. As has been reported even on this site, windows 8 will supposedly support ARM architecture. Time to buy ARM stock is now (well, ok, it was 18 months - 2 years ago, but still...).
  • ninjaquick - Wednesday, January 19, 2011 - link

    ARM designs are not anywhere complex or large enough to run against big die processors. ARM is a good foundation but they do not have any designs that can compete with Nehalem or Stars. I would love to ARM's take on mainstream computing, and its support in Windows 8 is exciting to say the least.
  • LiquidLearner - Wednesday, January 19, 2011 - link

    Perhaps current ARM designs aren't, but remember that ARM is a RISC based processor. The reason Macs used to be so much better at Photoshop and movie editing was because RISC is a more efficient architecture than x86. That's ultimately why they are able to use less power now, each instruction is less work than it is on x86. At least that seems to be my vague memory from classes 10 years ago. It was something one of the professors would complain about on a regular basis.

    If you have Windows 8 running on ARM and it is highly optimized for multi-core you could stack 16 Snapdragon cores (for example) into a single cpu and still use less power than most desktop CPUs currently use. All of these running at 1.5-2Ghz. That's an awful lot of processing power. This assumes that Windows 8 will be highly scalable to multiple cores which is likely a good assumption.

Log in

Don't have an account? Sign up now