What is the difference between DSP and Micro contoller?

[hysamir]

Just wanna know what make DSP chip special about doing FFT and many other thing.. and for what stuff DSP and MC are at best?


if we say DSP56k and MC68HC11 and our intel processor ?

 

[Naveed Alam]

DSP got speciality to be more efficient in mathematics..
MC is not that efficient in this regard

u can compare MC and MP.. MC has builtin MP and accessories..whereas MP is stand-alone MP only, having no accessories(RAM, ROM etc...)

 

[starfish]

DSPs are processors which are optimized for efficent calculation of many signal processing algos. A microcontroller is a general purpose IC which can be used in many applications.

Take the example of convolving input x [ n ] and impulse response h [ n ]

y[n] = x [ n ] * h [ n ]

and suppose that impulse response h [ n ] has 10 values and x [ n ] has 10 values. So from the convolution formula , in order to compute one sample of output y[n] u need to multiply 10 samples of impulse response with your signal samples and then add 10 times to get one sample of the output. In a microcontroller this would be done in a loop utilizing the sample multiplier and adder units. But in DSP processors you have more than 1 multipliers and adder units , which make execution much faster. In fact many of the computationally intensive algos can't run on most microcontrollers because they are not optimized for so much additions and multiplications but DSPs have this facility.

 

[bmwcoupe]

hi

each one has its own use...

micro controller its for simple calculations and more for control (i/o's)

dsp's purpuse is to make some complicated calculations

and another issue is a MIPS - million operations in second..........
and 1 more issue is a speed, DSP is much much faster clock defenition than mcu..



good luck.

 

[puviarasu]

MCU - used for handling i/o and sequencing and scheduling tasks (like a supervisor) it don’t have any major mathematical units contains standard ALU and some contains a math co processor. and the peripherals are directly controlled by the CPU

DSP - its the next advance form of MCU which is specially designed for data processing , it contains many mathematical unit on a single CPU and can perform operations simultaneously, and the peripherals can also perform function simultaneously and contains many dma channels compared to MCU and can perform multiple instructions on a single clock cycles and contains multiple busses and execution pipeline on a single ic.

 

[cschneider27]

DSPs have different memory architecture. MC most often use shared program and data memory, whereas a 56k uses a super harvard architecture with separate program and data memory. The 56k for example has a program and 2 data memories internally which all can be accessed at the same time. This way DSP algorithms can be executed much faster, since two operants and the next instruction can be fetched within one clock cycle.

The ALU is a bit different. Most DSP use fixed point now and offer saturation and speical rounding modes, which are usually needed for DSP algorithms.

DSPs do not have a MMU, but in the case of a 56k they have special addressing modes, as for example automatic wrapping and special modulo modes, which match nicely with DSP algorithms.

HTH.