NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Round-off
From: John Karl
Date: 2009 May 16, 06:29 -0700
From: John Karl
Date: 2009 May 16, 06:29 -0700
Peter, I repeat: I am NOT talking about observation error from rolling decks, systematic error, plotting error, or anything else. I'm ONLY addressing round off error in the calculation of a sight reduction -- a St. Hilaire reduction. I count 10 numbers used in such a reduction. If every number is off by 0.05' in the unfavorable direction, the result is 0.5' in error due to round off. This 0.5' is the MAXIUM possible error. It's not a statistical measure (such as an average error, probable error, max entropy estimate) from a uniform distribution of the 0.05' uncertainties in the numbers. Of course, using 1.0' precision in the same calculation, such as W.F. Buckley does, gives a Max error of 10'. As I mentioned earlier, the other calculation uncertainties arise from the Almanac data, and reduction tables if they are used. An appreciation of this calculation max error is important in several cases. One is when you want to evaluate Cnav accuracy by comparing with a very accurately know position. Then it's necessary to minimize the calculation error so that the remaining error can be attributed to observing skill, plus all the other factors you mentioned. In the 1' Buckley calculation, frequently these aren't sufficiently separated to allow a conclusion. JK --~--~---------~--~----~------------~-------~--~----~ Navigation List archive: www.fer3.com/arc To post, email NavList@fer3.com To , email NavList-@fer3.com -~----------~----~----~----~------~----~------~--~---
