1.3. Increasing Performance While Lowering Costs
The decision to add an FPGA to a new or existing embedded system may be driven by the desire to extend the usefulness of a common, low-cost microprocessor (and avoid introducing a higher-end, specialized processor such as a DSP chip) or to eliminate the need for custom ASIC processing functions. In cases where the throughput of an existing system must increase to handle higher resolutions or larger signal bandwidths, the required performance increases may be primarily computational (requiring a scaling of computational resources) or may require completely new approaches to resolve bandwidth issues.
Digital signal processing (DSP) applications are excellent examples of the types of problems that can be effectively addressed using high-density FPGAs. Implementing signal processing functions within an FPGA eliminates or reduces the need for an instruction-based processor. There has therefore been a steady rise in the use of FPGAs to handle functions that are traditionally the domain of DSP processors. System designers have been finding the cost/performance trade-offs tipping increasingly in favor of FPGA devices over high-performance DSP chips andperhaps most significantlywhen compared to the risks and up-front costs of a custom ASIC solution.
Most computationally intensive algorithms can be described using a relatively small amount of C source code, when compared to a hardware-level, equivalent description. The ability to quickly try out new algorithmic approaches from C-language models is therefore an important benefit of using a software-oriented approach to design. Reengineering low-level hardware designs, on the other hand, can be a tedious and error-prone process.
FPGAs help address this issue in two ways. First, they have the potential to implement high-performance applications as dedicated hardware without the up-front risk of custom ASICs. Mainstream, relatively low-cost FPGA devices now have the capacity and features necessary to support such applications. Second, and equally important, FPGAs are becoming dramatically easier to use due to advances in design tools. It is now possible to use multiple software-oriented, graphical, and language-based design methods as part of the FPGA design process.
When implementing signal processing or other computationally intensive applications, FPGAs may be used as prototyping vehicles with the goal of a later conversion to a dedicated ASIC or structured ASIC. Alternatively, FPGAs may be used as actual product platforms, in which case they offer unique benefits for field software upgrades and a compelling cost advantage for low- to medium-volume products.
High-capacity FPGAs also make sense for value engineering. In such cases, multiple devices (including processor peripherals and "glue" logic) may be consolidated into a single FPGA. While reduced size and system complexity are advantageous by-products of such consolidation, the primary benefit is lower cost. Using an FPGA as a catchall hardware platform is becoming common practice, but such efforts often ignore the benefits of using the FPGAs for primary processing as well as just using them for traditional hardware functions.
In many DSP applications, the best solution turns out to be a mixed processor design, in which the application's less-performance-critical components (including such elements as an operating system, network stack, and user interface) reside on a host microprocessor such as an ARM, PowerPC, or Pentium. More computationally intensive components reside in either a high-end DSP, in dedicated hardware in an FPGA, or in a custom ASIC as appropriate. It is not unusual, in fact, for such systems to include all three types of processing resources, allocated as needed. Developing such a system requires multiple tools and knowledge of hardware design methods and tools but provides the greatest benefit in terms of performance per unit cost.
For each processor type in the system (standard processor, DSP, or FPGA), there are different advantages, disadvantages, and levels of required design expertise to consider. For example, while DSPs are software-programmable and require a low initial investment in tools, they require some expertise in DSP-specific design techniques and often require assembly-level programming skills. FPGAs, on the other hand, require a relatively large investment in design time and tools expertise, particularly when hardware design languages are used as the primary design input method.
When compared to the expertise and tools investment required for custom ASIC design, however, FPGAs are clearly the lower-cost, lower-risk solution for developing custom hardware.
FPGAs provide additional advantages related to the design process. By using an FPGA throughout the development process, a design team can incrementally port and verify algorithms that have been previously prototyped in software. This can be done manually (by hand-translating C code to lower-level HDL), but C-based design tools such as those described in this book can speed this aspect of the design process.