What do you do when the CPU in a boardlevel system cannot cope with the software upgrades that the next-generation product inevitably requires? You can increase the clock speed, but when&#151not if&#151that approach runs out of steam, you must design in a faster CPU, or worse, an additional CPU. Worse yet, you could add another CPU board if the end product’s form factor and cost target allow it.

In this article, we’ll discuss some of the pros and cons of these approaches and describe a programmable coprocessor solution that leverages an on-board Xilinx FPGA to turbocharge the existing CPU and deliver the pros with none of the cons. You do not need processor design expertise, nor will you have to redevelop the software. Moreover, if the board already deploys an FPGA with sufficient spare capacity, the incremental silicon cost of the solution is zero.

Reprinted with permission from Xcell Journal / Third Quarter 2006. Article © Xcell Journal.