MCU math chip adds up.(Embedded Math)

ECN-Electronic Component News, August, 2006 by Titus, Jon

Although compilers and assemblers produce tight integer-math code, floating-point math can require memory-hogging code libraries that emulate math operations. So, a small MCU might stall when it must compute trig or exponential values.

Over the years, engineers have implemented math operations in hardware. In the mid '70s, I interfaced a Texas Instruments TMS0117 binary-coded decimal calculator chip to a microprocessor system to perform four-function math. Although a bit of a kludge, the circuit worked. In the 1980s IBM personal-computer motherboards included a socket for a "math co-processor" chip--the Intel 8087.

[ILLUSTRATION OMITTED]

Newer 16- and 32-bit processors provide internal math operations, address lots of memory or operate with...

Read the rest of this article with a Free Trial at HighBeam Research.

Premium Content Partnership | HighBeam Research provides an in-depth online archive library of reference works.