AMD Really Can't Wait to Launch This Important Product

It's been just more than a year since Advanced Micro Devices (NYSE: AMD  ) decided to build chips around platforms not inherited from PC processor giant Intel (NASDAQ: INTC  ) . Then we saw a road map for 2014 where AMD's chips based on ARM Holdings (NASDAQ: ARMH  ) technologies would sample in the first quarter of 2014 and go into full production in the second half of the year.

Well, AMD just put the pedal to the metal. The first AMD chip with ARM cores inside are scheduled for "imminent" sampling, and we're only one month into that first quarter.

AMD is aiming its ARM chips straight at the heart of traditional workloads for virtualized systems, like cloud computing and web servers. Image source: AMD marketing materials.

The product formerly known under the Seattle codename now has a proper title; the ARM-based server chips will be sold under the Opteron A moniker. The first such product, the Opteron A1100 series, will feature ARM Cortex A57 processor cores, running four or eight parallel cores at 2 GHz or more over a 64-bit system bus, coupled with server-class memory systems. Translation for the lesser geeks in the audience: These chips will be both faster and more reliable than your average ARM chip for smartphones and tablets.

The big selling point for these server chips, of course, won't be the straight-up number-crunching performance. Intel kind of has a lock on that market these days with its hard-to-match Haswell and upcoming Broadwell processors.

Instead, AMD is betting on extreme performance per watt. Data centers are bulging under the power and cooling needs of modern high-end processors, which is why it makes sense to import ultra-efficient designs from ARM's traditional mobile hunting grounds into this space.

These early ARM chips for the data center will bring "the experience and technology portfolio of an established server processor vendor to the ARM ecosystem and provides the ideal complement to our established AMD Opteron x86 server processors," said Suresh Gopalakrishnan, AMD's VP of server products.

What is AMD doing right this time?
That remains to be seen, of course. But the platform springs to life fully formed, matched with AMD-vetted development tools and an optimized version of Red Hat 's open-sourced Fedora Linux system. It's a good start, and AMD looks ready to hit the ground running.

The press materials still place the release date for the final product in the second half of 2014, but I'd be surprised if it doesn't show up over the summer. AMD is clearly putting its back into getting this product out the door as early as possible, and for good reason. The first-mover advantage could be crucial to AMD's success in the ARM-based server space, and heaven knows there's a ton of other ARM chip vendors ready to pounce on any weakness.

And Intel may have sold its ARM license to Marvell in 2006, but still hopes to capture the low-power server market with its own line of Atom processors -- tweaked to move out of laptops and tablets, into the data center. Intel is a rival not to be taken lightly, even if the Atom line hasn't been burning up the charts so far.

So the time to move on radically low-power server chips is yesterday, and it's good to see AMD stepping up its game here.

Rob Wall Street with these 3 disruptive growth stocks
The one sure way to get wealthy is to invest in a groundbreaking company that goes on to dominate a multibillion-dollar industry. Our analysts have done it before with tremendous returns on seemingly doomed investments. And now they think they've done it again with three stock picks that they believe could generate the same type of phenomenal returns. They've revealed these picks in a new free report that you can download instantly by clicking here now. No, AMD's turnaround story isn't one of them -- but you won't walk away disappointed from this report!


Read/Post Comments (7) | Recommend This Article (1)

Comments from our Foolish Readers

Help us keep this a respectfully Foolish area! This is a place for our readers to discuss, debate, and learn more about the Foolish investing topic you read about above. Help us keep it clean and safe. If you believe a comment is abusive or otherwise violates our Fool's Rules, please report it via the Report this Comment Report this Comment icon found on every comment.

  • Report this Comment On January 30, 2014, at 11:31 AM, techy46 wrote:

    AMD's been impatiently waiting to bring out their savior for the last 10 years and it never happens.

  • Report this Comment On January 31, 2014, at 11:16 AM, fearandgreed2005 wrote:

    Can you put some numbers on "extreme performance per watt"? I think Intel's performance per watt will still be far better than AMD's.

  • Report this Comment On January 31, 2014, at 11:50 AM, mtechac wrote:

    AMD is really successful when it brings ground breaking technologies. i.e. when they brought the 64-bit x86 instruction set. Intel almost lost its market.

    The HSA/hUMA architecture is 10 times bigger than the 64-bit x86 achievement because it sets AMD as the only company to have achieved the future of processors, that is, a hybrid CPU/GPU processor that schedules CPU and/or GPU cores, makes all cores share the same memory/process-space, locking, etc.

    Even though AMD GPU's are better than Intel's GPUs, Intel processors were always ahead of AMDs because all processing sequential or graphical have to be scheduled by the CPU, which means a faster CPU will always produce faster processing since graphics will have to wait.

    With the HSA/hUMA processors, that is not true anymore, The CPU and GPU can ran in parallel and the GPU don't wait for the CPUs any more, which means, many programs that are graphical in nature, will jump in performance significantly and will trump any Intel CPU.

    Although, the programs have to support HSA/hUMA programming, it is to expect since all future hybrid processors will not use the archaic programming models, which is why there are so many companies like Samsung, Texas Instruments, etc. that are the founders of the HSA/hUMA foundation.

    AMDs Kaveri is the first chip to be HSA/hUMA compliant, and current benchmarks just show the minimum performance side against Intel. When they run the real HSA/hUMA software, then AMDs HSA./hUMA chips will scale in performance. Those are the benchmarks that needs to be run on AMD Kaveri chips, otherwise, it will be like running a software without any optimizations of any kind and in the slowest cases possible.

    Anybody who understands CPU/GPU processors understand that what AMD has achieved is a new generation of processors that will change the way programming and processing is done.

    AMD can easily hike its performance with several additions to its architecture. i.e. RDDR5, etc. while the old Intel CPU architecture is at its limits and with serious bottlenecks that AMD has now solved.

    AMD can apply the same HSA/hUMA technologies to its ARM processors, which will make Intel chips look like Neanderthals..

    If Samsung will buy AMD, it will make Intel history. Samsung has excellent foundry processes that are very close to Intels. If Apple buys AMD, then Apple will be the most powerful computer company in world because it will own the finest computer processors design in history..

    Funny thing is that Apple created and heavily uses OpenCL and AMDs HSA/hUMA architecture is one of the best processors architectures for OpenCL.

    I think that the sky is the limit for AMD.. Rory Read is a great CEO and has surround himself with some of the finest people in the industry. Those people will not have joined AMD if they did not understand the technology and implications that AMD is starting to release...

  • Report this Comment On January 31, 2014, at 11:54 AM, bluesky64 wrote:

    Andrew,

    This will change the bottom line to AMD .

    Here's the latest mock up of earning for the first quarter earning 2014.

    http://www.scribd.com/doc/203457003/AMD-UNDERESTIMATED-GUIDA...

    This should make for a very strong yr.

  • Report this Comment On January 31, 2014, at 12:36 PM, rav55 wrote:

    @fearandgreed2005

    Numbers are pretty hard to come by when silicon is not available. Atom is not at all competitive with Arm in the mobile space, almost but not quite in the tablet space either. HD graphics are just appalling when stacked against Radeon or GeForce for that matter. x86 is just not as efficient as ARM.

    Since AMD will not be dependent upon Intel compilers to derate their silicon we should see very good performance from AMD.

  • Report this Comment On January 31, 2014, at 2:46 PM, wownwow wrote:

    Intel's mistake is squeezing AMD too much instead of letting AMD be a healthy (profitable) #2 to keep assiting it in x86.

    AMD was forced to find a capable CEO (Mr. Rory Read) and has been finding new lovers (markets) instead of sticking with the same old lover and is getting happier and happier.

  • Report this Comment On February 02, 2014, at 10:24 AM, alibabababa wrote:

    Really, a game changer? Well...sorry but Intel has new server chips coming out as well. As usual Intel will trump AMD. AMD had reported Q4 revenue of $1.59 billion. Intel had Q4 revenue of $13.8 billion. Does anyone really think this is a game changer? Hardly? AMD revenue growth for Q1 2014 is expected to decrease by16%. Even the company itself is telling you their outlook is bleak. At least Intel will be flat, but no decrease in revenue. These two companies need mobile. Remember this...Intel and photonics.

Add your comment.

DocumentId: 2816184, ~/Articles/ArticleHandler.aspx, 7/28/2014 2:55:32 AM

Report This Comment

Use this area to report a comment that you believe is in violation of the community guidelines. Our team will review the entry and take any appropriate action.

Sending report...


Advertisement