MIPS® DSP ASE

Digital Signal Processing on the Industry-Standard MIPS® Architecture (pdf)

DVD recorders, digital cameras, residential gateways and VoIP phones are examples of the growing list of consumer products that require an increasing amount of signal and media processing horsepower. In the cost-sensitive, high-volume consumer electronics market, eliminating unnecessary hardware and tool chains and reducing royalty payments can result in savings of millions of dollars.

To address these trends, MIPS Technologies has enhanced its MIPS32 industry-standard architecture with DSP functionality to provide a single design environment that leverages a common tool set and knowledge base. The MIPS DSP ASE offers licensees a programmable solution for DSP applications, allowing adaptation to changing market needs and extending the life of an SOC design. The new DSP extension comprises a set of new instructions and state in the integer pipeline of MIPS Technologies cores and requires less 5-10% additional silicon area to implement in a 24K-class core (depending on synthesis options).

DSP extension to the MIPS32® and MIPS64® architectures in the integer pipeline
- Integrates functionality
- Eliminates need for separate DSP for voice and audio (SD and HD) applications.
New instructions and new architectural state
- SIMD (8/16/32)
- Saturating fractional math
- MAC/dot-product
- Complex multiply support
- Variable bit insert
- Variable bit extract
- Virtual circular buffers
30 to 300% performance improvement for many applications
5-10% additional core area
Full development support

Revision 2 of the DSP ASE, available first in the 74K family of processor cores, provides additional support for image-processing and video-processing applications. It improves the orthogonality of the DSP ASE instruction set, making it an easier target for automatic compiler code generation. It also simplifies programming by introducing new instructions that perform more complex operations. Thus, using the DSP ASE Rev2 saves both development and execution time, as well as provides additional performance improvement for certain applications types like video and image processing.