You can also look at the post I made in this thread which speaks to the Ellis flow dynamic calibration method which is much better.
The one thing you want to do is to take a high res photo of the sample and zoom in on a larger screen monitor. It makes it much easier to tell which lines are better.