DALLAS, 9 Nov. 2010. Texas Instruments Inc. in Dallas is introducing the TMS320C66x digital signal processor (DSP) with fixed- and floating-point processing for aerospace and defense applications in radar, sonar, signals intelligence, and electronic warfare processing. TI also is introducing the TMS320C66x DSP devices, which include pin-compatible multicore DSPs in two-, four-, and eight-core versions -- the TMS320C6672, TMS320C6674 and TMS320C6678, and TMS320C6670.
"On floating-point performance the C66x delivered a BDTImark2000 score of 10,720," say analysts at Berkeley Design Technology Inc. (BDTI) in Oakland, Calif., an independent third-party analysis firm. "This will enable application developers to develop initial application implementations using floating-point math, and then decide whether performance-intensive sections of the code should be migrated to fixed-point to boost performance."
As a combined fixed-and floating-point DSP core capable of instruction by instruction execution of the two processing modes, TI's C66x core achieves even greater performance in an application where each segment of code is executed in its native processing mode. Integration of fixed- and floating-point capability in the C66x DSP core eliminates algorithm conversion from floating-point to fixed-point TI officials say.
The C66x DSP core was benchmarked at a clock speed of 1.25 GHz and achieved a BDTImark2000 score of 10,720 on the floating-point portion of BDTI DSP Kernel Benchmarks. For more information contact Texas Instruments online at www.ti.com.