Chipmaker AMD is giving us a glimpse of its future -- and it looks faster. The company on Wednesday unveiled its 2014
accelerated processing unit (APU) product roadmap at the APU13 Developer Conference.
Raising the performance bar across fanless tablets, 2-in-1s and ultrathin notebooks, the APUs codenamed “Mullins” and “Beema” are set to deliver more than twice the performance-per-watt of the previous generation.
“AMD is establishing excellent momentum this year in the low-power, mobile computing market and with ‘Mullins’ and ‘Beema’ coming in 2014 we are not standing still,” said Mark Papermaster, AMD’s chief technology officer and senior vice president, during his closing keynote at APU13. “AMD aims to deliver a set of platforms in the first half of next year that will outperform the competition in graphics and total compute performance in fanless tablets, 2-in-1s and ultrathin notebooks.”
Faster, More Secure
The latest AMD APUs also support InstantGo for faster wake times and to ensure like e-mail actively refresh in standby. Both of the new families are also the first to integrate an AMD-developed platform processor based on the ARM Cortex-A5 featuring ARM TrustZone technology for better data security.
Both new processor families offer two or four “Puma” CPU cores and AMD Radeon graphics on a 28nm system-on-chip. The new processors are planned to launch in the first half of next year and will be demonstrated at CES 2014 as part of a full suite of AMD products. The new low-power APUs join the previously disclosed high-performance notebook APU, codenamed “Kaveri,” in AMD’s 2014 mobile lineup.
We caught up with Charles King, principal analyst at Pund-IT, to get his take on AMD’s APU roadmap. He told us AMD is fighting two not entirely separate battles, at least in the consumer space. AMD is going head to head with Intel’s x86 technologies and competing with NVIDIA in the graphics processors space.
“ has done some significantly good work bumping up the performance of its native graphics capabilities that are included in the Core systems,” King said. “NVIDIA is taking a different direction. It’s still providing GPU technologies that can be added to Intel systems for gaming and high-performance graphics but it’s also attempting to do its own thing in the personal computing space by leveraging ARM and other technologies in standalone systems.”
In terms of AMD’s announcements this week, King finds it interesting that the company is pushing hard to leverage Opteron as the basis of systems that can be leveraged for VDI and other graphics-intensive workloads. He’s also taking note of AMD’s continuing drumbeat on the silicon vendor partnership front.
“The company has licensed ARM and is planning on migrating ARM technologies into its own platforms. But it’s doing a lot of work with companies like Qualcomm and Samsung and Texas Instruments in developing products and technologies that they can go to market with together,” King said.
“Those practical strategic partnership efforts are pretty singular for AMD. Engaging closer with ARM silicon partners is certainly not something Intel is pressing forward with on an aggressive basis. It will be interesting to see exactly what kind of work comes out of this and what kind of role AMD is able build and leverage for itself,” he added.