The new benchmark for wireless infrastructure DSP

The picoArray delivers substantially better performance per dollar than legacy DSP architectures, according to the BDTI Communications Benchmark (OFDM)™ benchmark.

Full results, based on a production-grade, volume-shipping PC102 device fabricated in 130nm technology, are shown below:

BDTI’s OFDM benchmark defines a “neutral” receive chain designed to allow cost and performance comparisons between many different types of chip. It includes a range of datapath-only functions such as Viterbi decoding, FFT, a 256-QAM demodulator, and FIR channel filter. The benchmark code is provided as plain C, which must be optimized by the processor manufacturer for maximum performance.

The benchmark calculates a total number of OFDM channels that may be implemented within a single device: and also a cost-per channel, based on the 1000-unit price of the device in question.

(Notes: Freescale result estimated; C6455 using Viterbi co-processor)

BDTI commented: “The 160 MHz PC102 is able to handle 14 channels of BDTI’s OFDM benchmark. The PC102 high-capacity results fall between those of the high-performance FPGA and the high-performance DSP processor, in terms of the number of channels supported and the associated cost per channel. The Texas Instruments high-performance ‘C6455 can only handle one channel of the benchmark, which means that the system designer would need to use multiple ‘C64x’s to implement a multi-channel application, or (more likely) a combination of 'C64x plus an FPGA. The Xilinx FX140, on the other hand, can handle many more channels than the PC102 The FX140 is much more expensive than the PC102, but has a lower cost-per-channel. (As mentioned earlier, the high-capacity results are optimized for maximum channels rather than minimum cost-per-channel.)

The low-cost results indicate that the PC102 again falls between the DSP processor and the FPGAs in terms of the number of channels supported and cost per channel.”

While the BDTI OFDM benchmark is neutral for datapath functions, it is worth noting that many real-world communications systems also demand a great deal of control plane processing. The picoArray, with its substantial general-purpose computing power and C-programmable architecture, can satisfy this requirement in a single chip solution: unlike FPGAs, which typically need to be used in combination with a supplementary programmable processor of some description.

Links:

• BDTI discussion of results
• BDTI benchmark results for all devices
• BDTI discussion of Picochip