A package based on an intensity-dependent normalization method for microarrays that is fast, simple and can include weighing of observations. TurboNorm is based on the P-spline scatterplot smoother using all data points for normalization, it uses invariant features without requirement to cover the complete intensity range. The method compensates for unequal coverage by using weights for all features, while the invariant features are given a higher weight. TurboNorm contains, besides a loess-like function for scatterplot smoothing, wrapper functions for both single and two-colour microarray data normalization. It also contains a function for adding the fitted curves on plots produced by the lattice package. The method has comparable properties to lowess and loess, but its lower computational complexity allows it to be faster and more memory efficient.