We take full advantage of the single instruction, multiple data (SIMD) capabilities of vector processors, and important routines are hand-tuned (often at the assembler level) to keep the pipeline running at full capacity.

VSIPL provides a portable, open standard interface to the library, but we also offer more direct access to the same power through our own C-language interface, CSIPL.

Our technology is used by a number of major DSP board manufacturers - why not join them? We can provide VSIPL and CSIPL for your boards If you want to specify your own API we can implement it easily.

The optimised library is currently available for

PowerPC 74xx with Altivec unit (G4), Linux or VxWorks
MIPS with paired-single and 3D extensions, Linux.

Multi-threaded versions for shared memory systems are also available.

FPGA

The Real-Time Systems Lab at QinetiQ has developed the Quixilica range of DSP cores for Xilinx FPGAs. One of these is a floating-point unit that can be used as a coprocessor to an embedded PowerPC. Alternatively, multiple FPUs with high bandwidth, low latency interconnections give performance that rivals dedicated DSP chips. N.A.Software is in partnership with QinetiQ to provide compilers and numerical libraries for the FPU.