John :
1.
Voila! Look at Fig 2, part 1 if you will!
There I demonstrated already in all details what you are suggesting!
2. etc.
I am somewhat frustrated that you did not answer my 2 simple questions I posed
in my "San Diego Convention" piece of yesterday. Without detailed answers to
detailed questions it is hard to communicate...
Look, John: it might quite well be that the theory you are describing is superior.
But as long as we don't have all the mechanics to apply it on board it
doesn't mean much to this group I think. We need tables, equations, drawings,
papers, various examples etc. that could be printed in Bowditch 2015.
Even if you did all that you would still have to demonstrate the superioriy of the
approach to make this work worthwhile. So let me I ask you if you permit:
Will it
improve accuracy, speed, reliability? Will it reduce perhaps the bookkeeping
and the calculations? Will it perhaps demand less from your brain as you try to handle
all the other challenges at sea? Are there efficient ways for checking results as you go?
Is it easy to memorize? Is it
straightforward to teach?
Are there special cases that are hard to handle?
Is there a somewhat reduced emergency solution?
Where would you get copies of necessary but lost documents?
What kind of
knowledge is required when you start learning it?
Or is it just a question of academic purity?
Best regards
h
From: John Karl <jhkarl---net>
To: hannoix---net
Sent: Tuesday, March 12, 2013 5:52 PM
Subject: [NavList] Re: The Darn Old Cocked Hat - the sequel 1
Hanno Ix,
1. I’m still not sure what you are calculating, or its significance (particularly your part 2). But in part 1, if you want the probability that the navigational observations performed at the true position (TL as you call it) yield a given distance from the TL, you should simple count the fractional number of your blue dots (compared to the total) inside each annulus divided by the area of that annulus. The dimensions of that result would be the probability per
unit area. Plot that versus the distance of the annulus from the TL. That’s what most navigators would be interested in – try it. And show us your plot.
2. As I’ve mention before, you’re starting with Gaussian (normal) distribution of lats and lons; I don’t know of any AstroNav sights that give us this.
3. And if you’re starting with fixes formed from two St. Hilaire LOPs, you’d need to calculated the lat & lon probabilities from the altitude probability distributions of the LOPs (there is no azimuth error). But this makes no sense either- we don't do astroNav this way.
Fair Winds,
JK
----------------------------------------------------------------
NavList message boards and member settings: www.fer3.com/NavList
Members may optionally receive posts by email.
To cancel email delivery, send a message to
NoMail[at]fer3.com
----------------------------------------------------------------
: http://fer3.com/arc/m2.aspx?i=122818
