Final Thoughts on AMD's Core Technology Update Presentation
Another Boring Presentation...?
In my old age I am turning into a bit of a skeptic. It is hard to really blame a guy; we are surrounded by marketing and hype, both from inside companies and from their fans. When I first started to listen in on AMD’s Core Innovation Update presentation, I was not expecting much. I figured it would be a rehash of the past year, more talk about Mullins/Beema, and some nice words about some of the upcoming Kaveri mobile products.
I was wrong.
AMD decided to give us a pretty interesting look at what they are hoping to accomplish in the next three years. It was not all that long ago that AMD was essentially considered road kill, and there was a lot of pessimism that Rory Read and Co. could turn AMD around. Now after a couple solid years of growth, a laser-like focus on product development based on the IP strengths of the company, and a pretty significant cut of the workforce, we are seeing an AMD that is vastly different from the one that Dirk Meyers was in charge of (or Hector Ruiz for that matter). Their view for the future takes a pretty significant turn from where AMD was even 8 years ago. x86 certainly has a future for AMD, but the full-scale adoption of the ARM architecture looks to be what finally differentiates this company from Intel.
Look, I’m Amphibious!
AMD is not amphibious. They are working on being ambidextrous. Their goal is not only to develop and sell x86 based processors, but also be a prime moving force in the ARM market. AMD has survived against a very large, well funded, and aggressive organization for the past 35 years. They believe their experience here can help them break into, and thrive within, the ARM marketplace. Their goals are not necessarily to be in every smartphone out there, but they are leveraging the ARM architecture to address high growth markets that have a lot of potential.
There are really two dominant architectures in the world with ARM and x86. They power the vast majority of computing devices around the world. Sure, we still have some Power and MIPS implementations, but they are dwarfed by the combined presence of x86 and ARM in modern devices. The flexibility of x86 allows it to scale from the extreme mobile up to the highest performing clusters. ARM also has the ability to scale in performance from handhelds up to the server world, but so far their introduction into servers and HPC solutions has been minimal to non-existent. This is an area that AMD hopes to change, but it will not happen overnight. A lot of infrastructure is needed to get ARM into that particular area. Ask Intel how long it took for x86 to gain a handhold in the lucrative server and workstation markets.
AMD has laid out year by year how they will address this new “ambidextrous” mentality and product alignment. One thing that stands above is really their “crown jewels” of IP; their x86 and graphics technologies. Their graphics IP is truly second to none and far exceeds that of what their primary rival (Intel) offers. It offers a source of differentiation for both x86 and ARM markets that their rivals so far have not been able to match. On the x86 side Intel does not have the flexibility, driver support, or overall performance of the latest GCN architecture on the Kaveri APU. On the ARM side there are several rivals with competitive IP (NVIDIA, Imagination Tech, ARM Mali), but none of these companies have working products that embrace a true heterogeneous architecture as AMD has already shown on the x86 side. Note that ARM and Imagination Technologies are partners in the HSA Foundation, so they will be supporting that functionality in the future. It just seems that AMD has a big head start so far on the field.
2014: The year AMD made contact with sustained profitability (or at least without significant losses). This will be a building block year for the company as they continue to leverage their existing Kaveri APUs into mobile markets as well as continuing with desktops and low end servers. Kaveri is their first true HSA enabled part that promises some significant compute performance with the right workloads and software support.
Beema and Mullins are optimized versions of last year’s Kabini and Temash APUs. These show a very significant improvement in power consumption and clockspeed, allowing these products to consume less power will still outperforming last year’s parts. These will be important products for AMD’s tablet and ultra-portable ambitions. It is not a great leap that AMD will also eventually implement these APUs into the AM1 platform as an upgrade to the current 25 watt Kabini products already there. A 15 watt APU that can clock significantly higher is a big win for AMD in this rather small, but still emerging market.
The big news is the availability of Seattle samples. Seattle is the codename for AMD’s first ARM based processor. Note I did not say APU. Seattle is aimed directly at the low power/blade server market. It does not contain any graphics IP, but it is comprised of up to 8 ARM Cortex A57 cores running at upwards of 2 GHz. This product will morph into the Opteron A1100. It will also include features such as an integrated DDR-3 and 4 memory controller, 8 lanes PCI-E 3.0, 8 SATA 6G ports, and perhaps most interestingly two 10 GbE connections. This product is only sampling, but will be available for purchase in late 2014 or early 2015. This is one of the more aggressive ARM server based offerings to be put up by a manufacturer.
Skybridge will allow either an ARM or x86 APU to be used on the same underlying hardware in 2015
There is only one thing missing from Seattle, and that is HSA. AMD is not at the point where they can integrate GCN or HSA functionality into the ARM Cortex A57 architecture (which is the design licensed directly from ARM). AMD showed off working silicon at their conference and it was able to do some fairly interesting workloads on the server side. We do not know yet how this architecture will scale or how it will handle exceptionally heavy loads. All we know is that it works and they are sampling it to partners.
Seattle is significant because it is the first non-x86 processor that AMD has offered in a long, long time. They licensed Intel designs when Intel couldn’t provide enough to the market, then extended that license and created their own x86 designs (and bought companies like NextGen). In the 80s and early 90s they produced their own RISC based designs that were contemporaneous with many SPARC based offerings. AMD is initially licensing the Cortex A57 core from ARM with the Seattle products, though it will not have a lot of AMD DNA integrated into the core architecture. The big moves here will be the memory controller and I/O options that AMD is designing into this product.