Freescale has launched a new family of application processors—the i.MX 6—that includes single-, dual-, and quad-core members along with a complement of hardware accelerators for multimedia applications. The processors combine ARM’s Cortex-A9 CPU with a 3-d graphics controller, video processing unit (VPU), image capture function, and image processing unit (IPU). The family targets a broad range of products from monochrome e-book readers and simple tablets at the low end to netbooks and full-function media tablets at the high end.

As we’ve all heard, the cost of creating a custom chip has skyrocketed. Still, there are applications where a custom chip is justified—usually by specialized, demanding technical requirements. Today, custom chips often incorporate multiple processor cores.

Ten years ago, I wrote about how licensable processor cores were beginning to play a more important role in the industry.  Among other trends, I observed that large chip companies were beginning to adopt licensable cores for application-specific chips such as cellphone baseband SoCs, rather than using proprietary cores that they developed in-house.  This trend has certainly strengthened over the past ten years.

Texas Instruments (TI) has announced four new ARM-based processors under the Sitara and Integra brand names. These processors are part of a pin- and software-compatible family that offers optional DSP and graphics accelerators to address applications with a wide range of performance requirements.

The term “the Internet of things” refers to a vision of the future where many objects—automobiles, appliances, sensors—will be connected to the Internet.

Let’s face it: Applications are getting more complicated.  Chips are getting more complicated.  And engineering teams are generally getting smaller, not larger.  As a result, it’s incumbent on chip vendors to provide robust, easy-to-use development kits.  Design engineers rely on these kits to quickly evaluate chips and prototype key portions of their systems.

In the last issue of Inside DSP, we explained why good benchmarks are important both for processor users and for processor designers.  We also discussed why optimized benchmarks are critical when assessing a processor’s performance for digital signal processing tasks.  Given that optimized benchmarks are required, if you’re a processor vendor, how do you ensure that your benchmark implementations are fully optimized, and therefore show your processor’s full potential?