With embedded hard digital signal processing (DSP) blocks, Stratix® series FPGAs are an ideal solution for high-performance, high-precision DSP applications. The DSP blocks are very power efficient and operate at far higher frequencies than the equivalent circuits in a soft implementation. When compared to DSP processors, Stratix series FPGAs deliver:
- DSP performance of an order of magnitude higher
- Ability to implement variable-precision signal processing (with Stratix V FPGAs only)
- Significantly lower system power consumption
- Smaller board space
- More platform scalability
The DSP block architecture has been optimized to implement various common DSP functions with maximum performance and minimum logic resource utilization. In addition to multipliers, each DSP block has functions that are frequently required in typical DSP algorithms. These functions include pre-adders, adders, subtractors, accumulators, coefficient register storage, and a summation unit. With these rich features, the DSP blocks in Stratix series FPGAs are ideal for applications with high-performance and computationally intensive signal processing functions, such as finite impulse response (FIR) filtering, fast Fourier transforms (FFTs), digital up/down conversion, high-definition (HD) video processing, and HD CODECs.
Table 1 compares the DSP resources in the Stratix series FPGA families.