ARM: PlayStation Graphics Power Coming To Smartphones
ARM demos at Mobile World Congress reveal smarter, more efficient processing that will let phones tackle tasks such as photo editing, too.
ARM is the power behind the mobile throne--not to mention washing machines, apparently. (It's true: ARM chips also power many mobile infrastructure components.) So it's always interesting to see what ARM has up its, ahem, sleeve at Mobile World Congress, especially a Mobile World Congress that witnessed an onslaught of quad-core phones that mostly use ARM, not to mention a Windows 8 demo on an NVideo Tegra3 reference tablet.
No, ARM wasn't talking about eight cores at this year's show. Apparently all of the work being done is around smarter, more efficient processing; that is, building in the intelligence so that all cores aren't necessarily working together like the cylinders of an auto engine. Instead they work at just the right time, on just the right task, and in doing so both make the phone faster and last longer on one charge.
ARM's lead mobile strategist, James Bruce, didn't say much about Intel's dive into the world of mobile. He only talked about ARM's push toward more power-efficient architectures.
Following are the highlights, some of which came by way of demonstrations at the show, which we've put in a video at the end of this article:
-- ARM has been working hard on mass-market smartphones, which is to say they want to help make smartphones cheap enough that "feature phones" become a thing of the past. ARM's Bruce mentioned Broadcom and Qualcomm specifically as providers of chipsets for low-cost devices. So if there are "feature phones" at the bottom end, and "mass market smartphones" in the middle, all of these new quad-core (and other) devices are "super phones."
-- ARM demonstrated its next-generation architecture for 2013, which it calls Big Little (which sounds a bit like a line from Dr. Seuss). Some tasks, Bruce noted, require only minimal power, but last for a long time (such as email and social media), whereas others require massive power. The next-generation architecture will be intelligent enough to detect when more power is needed and shift cores from "little" to "big" and vice versa.
-- Bruce also demonstrated how far we've come, running a benchmark that showcased Web rendering time using an ARM 11 (running on an original Android G1 device), the Cortex A8 running at 650 MHz (on a Motorola Droid), and the dual core A9 (running on a Galaxy Nexus). The benchmark showed a 10x improvement over the 3-1/2-year span of these phones (13 seconds on the G1, 1.2 seconds on the dual core A9). He also showed an HTC phone (a "mass market" smart phone) running the Cortex A5, rendering the page in two seconds, again demonstrating how cheaper phones are becoming more powerful.
-- Finally, Bruce showed off the Mali graphics processor on a Samsung Galaxy S2. The reflection and lighting effects and complex textures, he said, were more like playing on a "PlayStation 2.5" gaming console--in other words, better than a PS2 and not quite up to PS3 quality. Soon enough, he said, PS3 quality will be available on a smartphone. More important for everyone, though, is that instead of just using the GPU to render the user interface, phones also will use it for heavy computing tasks such as photo editing.
InformationWeek Elite 100Our data shows these innovators using digital technology in two key areas: providing better products and cutting costs. Almost half of them expect to introduce a new IT-led product this year, and 46% are using technology to make business processes more efficient.
The UC Infrastructure TrapWorries about subpar networks tanking unified communications programs could be valid: Thirty-one percent of respondents have rolled capabilities out to less than 10% of users vs. 21% delivering UC to 76% or more. Is low uptake a result of strained infrastructures delivering poor performance?
InformationWeek Tech Digest, Nov. 10, 2014Just 30% of respondents to our new survey say their companies are very or extremely effective at identifying critical data and analyzing it to make decisions, down from 42% in 2013. What gives?