Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
As in wikipedia:
DSP's are mainly based on Harvard architecture to perform task faster with hard wired instructions as compared to Microcontrollers which are mainly available with von neumann architecture (some microcontrollers like PIC is an exception).
The DSP can compute the numerical parts of your application faster than the microcontroller can. Because that's what it's designed to do. You might also find that DSPs have more of the "right sort" of integrated peripherals, compared to microcontrollers.
There are many Harvard microcontrollers, DSPs are mostly Harvard, but the key advantages of DSP are:
- most instuctions are single cycle, and 16 or 32 bit devices;
- supports special instructions for signal processing: multiplications, multiply and accumulate the result (MAC) in 1 cycle, data adjusting, representation conversion;
- special addressing modes, circular buffers needed for filter implementation along with MAC, bit reversed addressing needed for FFT;
- DMA.
A few uC are also singlecycle and are 16b, but without the DSP features it takes a long time to move data back and forth in software, so they can't be used as DSP for a resonable complex project.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
