Re: AstronomicalData and assumed geographic position
- To: mathgroup at smc.vnet.net
- Subject: [mg94192] Re: AstronomicalData and assumed geographic position
- From: jmbryant12 at gmail.com
- Date: Sat, 6 Dec 2008 19:57:47 -0500 (EST)
- References: <ghdmsj$fo7$1@smc.vnet.net>
Thanks for noticing the issue. There was indeed a problem with the
non latitude/longitude specified form. I've narrowed this down and
should have a fixed version by early next week if all goes well. The
nice thing about these data functions is that they are not necessarily
tied to a specific release and so problems like this can be fixed
quickly. I'll need to get the fix implemented and rebuilt and should
be able to get it tested early next week. Once approved, you should be
able to re-evaluate in a fresh kernel and you'll get the fixed
behavior.
As for speed, the following takes just less than 8 seconds on my
somewhat slow home machine:
Plot[AstronomicalData[
"Sun", {"Altitude", {2008, 12, 5, h , 0, 0}, {40.11, -88.24}}], {h,
0, 23}]
This is querying AstronomicalData 278 times, one for each sampling
point in the graphic. One individual call takes only .031 seconds on
my machine so that's pretty fast. Plot does automatic sampling of this
call and results in more calls than you probably thought were being
made. For example, by turning off recursion in the Plot, it now takes
slightly over 1 second on my machine:
Plot[AstronomicalData[
"Sun", {"Altitude", {2008, 12, 5, h , 0, 0}, {40.11, -88.24}}], {h,
0, 23}, MaxRecursion -> 0]
I can speed it up further by reducing the default sampling rate which
takes about .3 seconds on my machine.
Plot[AstronomicalData[
"Sun", {"Altitude", {2008, 12, 5, h , 0, 0}, {40.11, -88.24}}], {h,
0, 23}, MaxRecursion -> 0, PlotPoints -> 10]
Its not AstronomicalData that is slow, but what you are doing with it.
Hope this helps. Thanks again,
-Jeff Bryant
Wolfram Research, Inc.
On Dec 6, 5:15 am, "sjoerd.c.devr... at gmail.com"
<sjoerd.c.devr... at gmail.com> wrote:
> Hi folks,
>
> According to the documentation for AstronomicalData "The default
> location on Earth to assume is given by the value of $GeoLocation."
>
> This means that
> AstronomicalData["Sun",{"Altitude",{2008,12,5,13,0,0}] and
> AstronomicalData["Sun",{"Altitude",{2008,12,5,13,0,0},$GeoLocation}]
>
> should yield the same results. In fact, they do not.
>
> Moreover, plotting one daily cycle with
>
> Plot[AstronomicalData[ "Sun", {"Altitude", {2008, 12, 5, h 0, 0}}],
> {h, 0, 23}] suggests a completly static sun.
>
> Therefore, it seems that AstronomicalData without explicit lat,long
> coordinates does not work properly. Or am I doing something wrong?
>
> By the way, it's a pity there is no built-in sun path function. Using
> AstronomicalData to get the sun position is about 20 times slower than
> using the VSB functions by Greg Pelletier (http://www.ecy.wa.gov/
> programs/eap/models/twilight.zip) that I converted to Mathematica
> code. AstronomicalData seems to be less accurate as well, but that may
> be due to differences in modelling of atmospheric refraction.
>
> Cheers -- Sjoerd