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
Comments Locked

177 Comments

View All Comments

  • Shadowmaster625 - Monday, June 20, 2011 - link

    Can you try setting the affinity manually so that it has to use only one core? I notice that even though I'm running only one thread, that thread will jump around to all my available cores, making them all look like they are running at 25%. Maybe by the time AMD's turbo kicks in, the thread is already moved to another core.
  • GullLars - Tuesday, June 14, 2011 - link

    Looking forward to the re-test with an SSD.
    IMHO all machines over $600 meant for general purpose use in 2011 should include at least a 32GB SSD of Indilinx Barefoot performance or better.
    My laptop from 2007 with 2,2GHz C2D and a 32GB Vertex (retrofitted in 2009) still wipes the floor with new $800+ laptops with HDD-only for general use.
  • JarredWalton - Tuesday, June 14, 2011 - link

    Having played with a bunch of laptops using 64GB SSDs, they all feel snappier, though of course it doesn't help with gaming frame rates or CPU computations. Still, I have to say that 64GB isn't big enough for me. If you can get a 32GB mSATA SSD and some sort of SSD caching, and then have a main 500GB HDD, that would be the sweet spot. If you're going SSD-only, I need at least 120GB, and for anything that can run games I'd want 240GB. (By the time I install most of the games I'm currently interested in playing on a regular basis, I easily fill up a 120GB drive.)

    Anyway, I'm swapping in the SSD now and will start testing during the week, with the follow-up article hopefully ready next week. I've got DC SNB, Arrandale, DC Phenom II, E-350, and now Llano for the article. Anything else you want to request before I call it quits? I've still got the XPS L502x and an AMD K625, but I figure the five laptops are a reasonable representation of what's currently out there. (Note that I have currently focused only on IGP equipped laptops.)
  • Boissez - Tuesday, June 14, 2011 - link

    Results with/without DDR3 RAM would be interesting as it seems RAM speed is a bottleneck.
  • Boissez - Tuesday, June 14, 2011 - link

    dang... That should have been with/without DDR3 1600 Mhz RAM
  • Brian23 - Tuesday, June 14, 2011 - link

    Application compile benchmarks. ;-)
  • Cloudie - Tuesday, June 14, 2011 - link

    The battery life has exceeded my expectations although battery life is a low priority for me when buying a laptop. As long as it gets 4 hours or so at idle is fine by me.

    Both the GPU and CPU performance has disappointed me somewhat but on the plus side I was not aware until now that the APU being tested was a 35W version, I simply assumed it was the fastest 45W one. Hopefully the CPU perfomance will be just that little bit better on the 45W APUs. And OEMs better get out some decent compact systems with this in... I'd say 13-14" chassis will do me just fine.
  • just4U - Wednesday, June 15, 2011 - link

    If they can get this priced out in a $600 laptop with all the bells and whistles... It will sell. Looking locally (and online) decent gaming laptops start out at about $800. Looking at the numbers it's more then acceptable I think. Profit has to be in there though for AMD.. If their not turning an acceptable profit on each chip then... it will be a wash.
  • duploxxx - Wednesday, June 15, 2011 - link

    Looking at the recent preview of anandtech on the desktop Liano part and the increased performance from DDR1333 to DDR 1866 i would like to see what it would bring on a Mobile platform. Afterall LIano mobile does support ddr 1600

    http://www.anandtech.com/show/4448/amd-llano-deskt...
  • Germanicus - Wednesday, June 15, 2011 - link

    1600 MHz DDR3 - lets make this accurate, since Llano can utilize it.

    Thank you.

Log in

Don't have an account? Sign up now