4 ms·
I remember in high school precalc, we learned about the Taylor series, and my teacher told us that it was how trig functions were actually implemented on calcul
by rmorey 2y ago
I remember in high school precalc, we learned about the Taylor series, and my teacher told us that it was how trig functions were actually implemented on calculators. Well I looked it up and found that it was actually CORDIC, and went and had some fun implementing it in TI Basic
- rossant 2y agoWell, are there _any_ calculators out there that use Taylor expansions?
- snovv_crash 2y agoI've done some work with fast approximate math libraries,and went down this path. Taylor was a dead-end, but Chebyshev polynomials work great. They have the nice property of having (close to) the minimum maximum (minimax) error over the entire range you are approximating.
- toolslive 2y agowell, from memory (this was > 20y ago) - Taylor gives you polynomial optimized for the evaluation of one point. - naive Newton gives you a polynomial with 0 error in the interpolation points [but the system has a bad condition] - Chebychev gives you a polynomial with minimal error over the interval of choice. So there is no real reason to ever use the Taylor series to approximate a function over an interval. The high school math teachers are lying.
- rmorey 2y agonot lying, probably just mistaken
- djmips 2y agoI thought you might be interested to read how the remarkable Sinclair scientific calculator did it's trig, log etc. It wasn't Cordic but the algorithm has similarities. http://files.righto.com/calculator/sinclair_scientific_simulator.html?m=1 http://files.righto.com/calculator/sinclair_scientific_simul...