NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: HO 211 with Sadler method
From: Paul Hirose
Date: 2016 Jul 26, 13:05 -0700
From: Paul Hirose
Date: 2016 Jul 26, 13:05 -0700
On 2016-07-25 22:33, Robert VanderPol II wrote: > The value I would think most important would be max error. Where would you draw the "unacceptable" line? I have started to write up some conclusions, based on a max of 5', but I'd like to hear what others think. Note that an absolute limit at 5' or any other value is maybe not realistic. To get the max error I use a run of one million tests, since the value jumps around noticeably with only 100,000 tests. In other words, the really bad Ageton reductions occur in highly specific conditions and are statistically rare. And we'd normally detect those sights from the appearance of the plot. For instance, vanilla Ageton has a max error around 32', but 99.3% of sights are within 5'. If we stipulate that, say, that no more than 0.1% exceed a certain threshold, that may be a more meaningful standard to shoot for.
