# Is n or nlog(n) better than constant or logarithmic time

algorithmtime-complexity

In the Princeton tutorial on Coursera the lecturer explains the common order-of-growth functions that are encountered. He says that linear and linearithmic running times are "what we strive" for and his reasoning was that as the input size increases so too does the running time. I think this is where he made a mistake because I have previously heard him refer to a linear order-of-growth as unsatisfactory for an efficient algorithm.

While he was speaking he also showed a chart that plotted the different running times – constant and logarithmic running times looked to be more efficient. So was this a mistake or is this true?

#### Best Solution

It is a mistake when taken in the context that O(n) and O(n log n) functions have better complexity than O(1) and O(log n) functions. When looking typical cases of complexity in big O notation:

O(1) < O(log n) < O(n) < O(n log n) < O(n^2)

Notice that this doesn't necessarily mean that they will always be better performance-wise - we could have an O(1) function that takes a long time to execute even though its complexity is unaffected by element count. Such a function would look better in big O notation than an O(log n) function, but could actually perform worse in practice.

Generally speaking: a function with lower complexity (in big O notation) will outperform a function with greater complexity (in big O notation) when n is sufficiently high.