NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Computer space required to get 1" accuracy
From: Ken Muldrew
Date: 2010 Apr 09, 09:24 -0600
From: Ken Muldrew
Date: 2010 Apr 09, 09:24 -0600
On 8 Apr 2010 at 23:28, Antoine Couette wrote: > I interpreted the "challenge" from Frank as follows : from a data array > to be defined as "small" as possible, use only brute force LinInt as a > "reason" (or an "excuse") to compute as fast as possible (therefore > Chebishev not a valid method here). Yes, I have the same interpretation. I just meant that if you could determine the relationship between the number of Chebychev coefficients and the divergence from linearity then you could calculate the size of your look-up table without actually writing the program. For my own part, when it was my turn to write an almanac-generating program I just used someone else's freely available container. My particular eccentricity was to create an almanac with current data but presented in the style of the old Nautical Almanacs. You can see a sample output here: http://people.ucalgary.ca/~kmuldrew/celestial/almanac_11_06.PDF And the source code, though uninteresting, is here: http://people.ucalgary.ca/~kmuldrew/celestial/almanac.html Ken Muldrew.