A small lamp of power radiates light uniformly in all directions

PHYSICS · Subjective · Fermi University Physics · Mathematical Methods for Physics · Numerical and computational mathematics · Numerical and computational mathematics

Computational Physics, Ch Numbers a machine can hold, and the two errors that follow

Problem

A small lamp of power P=150 WP = 150\,\mathrm{W} radiates light uniformly in all directions, giving an intensity I(d)=P/(4πd2)I(d) = P/(4\pi d^2) at distance dd from the lamp. The distance is measured as d=3.20 md = 3.20\,\mathrm{m} with a relative uncertainty of 0.80%0.80\%. (a) Compute the condition number κ=∣df′(d)/f(d)∣\kappa = |d f'(d)/f(d)| of II as a function of dd. (b) Use κ\kappa to bound the relative uncertainty in the computed intensity I(d)I(d), and quote I(d)I(d) with its absolute uncertainty. (c) Suppose N=1.00×104N = 1.00\times10^4 independent readings, each equal to I(d)I(d) to within rounding, are summed in a counted loop as part of a longer calculation. Using the worst-case rounding bound 12Nϵmach\tfrac12 N\epsilon_{\rm mach}, give the resulting relative error from summation alone, and say whether it can compete with the uncertainty found in part (ii).

Browse the problem library