However compelling a new product may be, it won’t succeed unless prospective customers know about it. Achieving customer awareness can be particularly challenging for small companies that lack multi-million-dollar marketing budgets, but it can also pose a challenge for established companies entering new markets.

These days, there are so many start-ups developing programmable processors that it feels like we’re back in the “bubble” years, when anyone with a remotely viable processor design could secure venture funding. A pivotal question for the current crop of start-ups is whether to offer their processors as flexible, general-purpose chips, or as highly specialized, application-specific solutions. Should it be a jack-of-all-trades, or a master of one?

This month BDTI and silicon intellectual property licensor ARC International announced completion of BDTI Solution Certification™ of the H.264 video decode performance of the ARC Video Subsystem.  The ARC Video Subsystem, the first product to be certified under BDTI’s Solution Certification Service, is a programmable subsystem capable of supporting multiple video standards.  In certifying the solution, BDTI has independently verified its performance using proprietary BDTI bitstreams and metrics.

Stream Processors, Inc. (SPI) this week unveiled its data-parallel processor architecture and announced two chips based on the architecture.  According to SPI, its architecture is optimized for compute-intensive embedded applications which exhibit a high degree of data parallelism, such as video and imaging.  SPI believes that cost-performance and developer productivity advantages will enable its chips to compete successfully against FPGAs, high-end DSPs, and ASICs in these applications.

Algorithms are the essence of digital signal processing; they are the mathematical “recipes” that transform signals in useful ways. Companies developing new DSP algorithms, or considering purchasing or licensing algorithms, often need to assess whether the algorithm will fit within their processing budget–and thereby within their cost and power consumption targets. 

At the Consumer Electronics Show last month I was struck (not for the first time) by the number of consumer electronics products that rely on digital signal processing—at this point, nearly all of them. In fact, so many of today's products incorporate digital signal processing-based functions that it's tempting to start viewing these functions as commodities.

Measuring the performance of real-time digital signal processing code is essential.  But whether you're using a simulator or hardware, it can be a headache to get accurate, repeatable performance measurements.  In this article we'll cover some of the common pitfalls you might encounter, and present some techniques for working around them.