BDTI COMMUNICATIONS BENCHMARK (OFDM)™ RESULTS
BDTI
HOME << FREE INFO << BDTI


BDTI Communications Benchmark (OFDM)™ Results


About the Results

The BDTI Communications Benchmark (OFDM)™ is an application-oriented benchmark based on an orthogonal frequency division multiplexing (OFDM) receiver. It is designed to be representative of the processing found in communications equipment for applications such as DSL, cable modems, and wireless systems.

The BDTI Communications Benchmark (OFDM)™ is used to evaluate and compare the performance of DSPs, general-purpose processors, multi-core processors, and FPGAs, among other types of processing engines. This benchmark provides two classes of results: high-capacity (maximum channels) results, and low-cost (maximum channels/dollar) results. Vendors may use different benchmark implementations and different chips to generate these two results.

To enable quick, realistic comparisons between chips, BDTI publishes the high-capacity (maximum BDTIchannels) and low-cost (BDTIchannels/$) metrics free of charge. The BDTIchannels/$ score is  based on a chip's low-cost implementation of the BDTI Communications Benchmark (OFDM). A higher score indicates better cost-performance (i.e., more BDTIchannels supported per dollar).  The maximum BDTIchannels score is based on a chip's high-capacity implementation of the BDTI Communications Benchmark (OFDM) .  As with the BDTIchannels/$ score, a higher score indicates higher performance.


Detailed benchmark results for DSP-enhanced FPGAs and DSPsincluding high-capacity and low-cost resultsare available in BDTI's new report, FPGAs for DSP, Second Edition.

View the Latest BDTI-Certified Results



ofdm_channel_dollar
  1. The definitions of speed grades vary among vendors. Therefore, a -5 speed grade for a Xilinx chip is not necessarily equivalent to a -5 speed grade for an Altera chip. For more information on speed grades visit the respective vendor websites.
  2. Estimated performance based on related benchmark implementations.

ofdm_channels_high_capacity
                               
  1. Without using Viterbi coprocessor.
  2. Using Viterbi coprocessor.
  3. Estimated performance based on related benchmark implementations.





Top of page