One of my first year students, Dan, is performing an investigation into the relationship between the amount of water in a wine glass and the frequency of sound it emits when stroked. He used Audacity to sample the sounds and produced frequency spectra for a range of depths but found no difference even though the difference is clearly audible. It seems to be a problem with recording multiple runs on the same file (maybe). As a side experiment he compared the same sound using two computers and found that there was a considerable (4 Hz) difference between the results. Could this be due to the frequency response of the different microphones? Interesting.