Want to increase your team's productivity? Reduce Bugs? Meet deadlines? Take Jack's one day Better Firmware Faster seminar. You’ll learn how to estimate a schedule accurately, thwart schedule-killing bugs, manage reuse, build predictable real-time code, better ways to deal with uniquely embedded problems like reentrancy, heaps, stacks and hardware drivers, and much, much more.

Jack will be presenting this seminar in Chicago (April 23, 2008), Denver (April 25) and London, UK (May 19). Want to be your company’s embedded guru? Join us! More info here.

For hints, tricks and ideas about better ways to build embedded systems, subscribe to The Embedded Muse, a free biweekly e- newsletter. No advertising, just down to earth embedded talk.

Click here to subscribe.

Floating Point Approximations
Often we must create our own versions of trig and other functions. Perhaps the compiler's floating point package is too slow, or lacks precision. Sometimes it's so huge that doing a lousy cosine doubles the program's size. Regardless, there are lots of options other than relying on what the compiler provides.

This paper gives a number of simple floating point trig approximations. The code includes test cases to check for maximum error. This paper gives approximations for other functions like roots, exponentiation and logs.

Back to the Ganssle Group's home page.

The Ganssle Group 
PO Box 38346, Baltimore, MD 21231 
Tel: 410-504-6660, Fax: 647-439-1454
Email info@ganssle.com 
© 2008 The Ganssle Group