What Took So Long?

AMD announced the acquisition of ATI in 2006. By 2007 AMD had a plan for CPU/GPU integration and it looked like this. The red blocks in the diagram below were GPUs, the green blocks were CPUs. Stage 1 was supposed to be dumb integration of the two (putting a CPU and GPU on the same die). The original plan called for AMD to release the first Fusion APU to come out sometime in 2008—2009. Of course that didn't happen.

Brazos, AMD's very first Fusion platform, came out in Q4 of last year. At best AMD was two years behind schedule, at worst three. So what happened?

AMD and ATI both knew that designing CPUs and GPUs were incredibly different. CPUs, at least for AMD back then, were built on a five year architecture cadence. Designers used tons of custom logic and hand layout in order to optimize for clock speed. In a general purpose microprocessor instruction latency is everything, so optimizing to lower latency wherever possible was top priority.

GPUs on the other hand come from a very different world. Drastically new architectures ship every two years, with major introductions made yearly. Very little custom logic is employed in GPU design by comparison; the architectures are highly synthesizable. Clock speed is important but it's not the end all be all. GPUs get their performance from being massively parallel, and you can always hide latency with a wide enough machine (and a parallel workload to take advantage of it).

The manufacturing strategy is also very different. Remember that at the time of the ATI acquisition, only ATI was a fabless semiconductor—AMD still owned its own fabs. ATI was used to building chips at TSMC, while AMD was fabbing everything in Dresden at what would eventually become GlobalFoundries. While the folks at GlobalFoundries have done their best to make their libraries portable for existing TSMC customers, it's not as simple as showing up with a chip design and having it work on the first go.

As much sense as AMD made when it talked about the acquisition, the two companies that came together in 2006 couldn't have been more different. The past five years have really been spent trying to make the two work together both as organizations as well as architectures.

The result really holds a lot of potential and hope for the new, unified AMD. The CPU folks learn from the GPU folks and vice versa. Let's start with APU refresh cycles. AMD CPU architectures were updated once every four or five years (K7 1999, K8 2003, K10 2007) while ATI GPUs received substantial updates yearly. The GPU folks won this battle as all AMD APUs are now built on a yearly cadence.

Chip design is also now more GPU inspired. With a yearly design cadence there's a greater focus on building easily synthesizable chips. Time to design and manufacture goes down, but so do maximum clock speeds. Given how important clock speed can be to the x86 side of the business, AMD is going to be taking more of a hybrid approach where some elements of APU designs are built the old GPU way while others use custom logic and more CPU-like layout flows.

The past few years have been very difficult for AMD but we're at the beginning of what may be a brand new company. Without the burden of expensive fabs and with the combined knowledge of two great chip companies, the new AMD has a chance but it also has a very long road ahead. Brazos was the first hint of success along that road and today we have the second. Her name is Llano.

The Llano A-Series APU
POST A COMMENT

177 Comments

View All Comments

  • duploxxx - Sunday, June 19, 2011 - link

    talking about crap??? men you are good at that, as if you need a 2600K for some compiling and add a 6990 to an e-350.

    men you do know someting about computers :) your a joke
    Reply
  • BushLin - Wednesday, June 22, 2011 - link

    If you're going to accuse someone of talking crap (correctly or otherwise) it helps if you know the difference between you're and your when attempting to insult them. Reply
  • Broheim - Monday, June 27, 2011 - link

    I use a fast CPU for compiling because I actually like being productive rather than just staring at the screen.

    my point about the 6990 was that a faster GPU != a better user experience for the vast majority of users, but logic seems to be lost on you.
    Reply
  • Regenweald - Tuesday, June 14, 2011 - link

    I'm currently playing the Witcher extended with a 2Ghz X2 ,a 4670 and 800 Mhz memory on a desktop with no complaints. In game settings medium and high. Are you saying that 2 more tweaked 32nm cores, 80 more shaders and ddr3 1600 or 1800 memory will not offer a good mobile gaming experience ? please. Reply
  • jollyjugg - Wednesday, June 15, 2011 - link

    What kind of super computing application are you going to do in your laptop that you would need that "tremendous CPU power" that you are talking about. As somebody who has used both intel and amd machines for years I can tell you that for most user applications, you will hardly notice any difference in performance. The main complaint was that AMD machines were running hot particularly after intel cameup with power gating in nehalem in 2008. With this part you get a machine that runs way way cooler and almost 150- 200 bucks cheaper than comparable intel machines plus you get discrete quality graphics for free. Nobody can change cynics like you. Because you cant expect Intel to sell anything cheap you would want to AMD to sell things cheaper. Well if you want good things in life you should be prepared to pay. Dont write trash. Like gaming is not important for average user, tremendous computing power and 3 GHz CPU speed is also not important for the average PC user. But multimedia and movie rendering etc is. So go Llano!!! Reply
  • Seikent - Tuesday, June 14, 2011 - link

    This platform offer some good things, but if the prices aren't low it won't go well. Now it's quite easy to find a sandy bridge notebook with discrete gpu for a few more dollars.

    If ACF does achieve to work like CF in the future, it would be great!
    Reply
  • SteelCity1981 - Tuesday, June 14, 2011 - link

    Finally we have an intergrated graphics solution that's worth really talking about. Reply
  • aegisofrime - Tuesday, June 14, 2011 - link

    Judging by the performance, this part won't be priced very high. And that's a worry, considering that it packs 1.45 billion transistors. In comparison, Sandy Bridge is 995 million transistors, and sells for more money. Profit margins are gonna be tight on this one. Reply
  • JarredWalton - Tuesday, June 14, 2011 - link

    AMD has a slide that points out that compared to their previous generation, they're packing a 66 mm2 Northbridge, 200 mm2 CPU, and 1080 mm2 dGPU into a 228 mm2 package. They've made money this past quarter, so this should do better than Athlon/Phenom II. Reply
  • JarredWalton - Tuesday, June 14, 2011 - link

    Note: that's supposed to be 108 mm2 GPU, not 1080. Whoops. Reply

Log in

Don't have an account? Sign up now