bug report: Find Zero Crossings doesn't work properly

(I tried sending this to the mailing list, but it didn't appear in the archive, so I'm not sure it was received, so now I'm trying to post it to the forum.)

Find Zero Crossings doesn't work properly.  I checked this by fully zooming the waveform horizontally and vertically so that I can see every data point.  The beginning and ending points of the selection after Find Zero Crossings might be as far off as +/-0.01 (out of 1) on the vertical scale, and they are not necessarily on a positive slope.  Repeating the Find Zero Crossings produces unpredictable results and sometimes results in a change in the selection and sometimes doesn't.  This occurs on both 1.2.6 and 1.3.3 beta.  I'm using Mac OS X 10.4.10, but the problem has also occurred on older versions of Mac OS X 10.4.

--Menasheh