Spock wrote:
> 
> I hate to keep beating on this point, but I still think the geometry is
> confused.

	First off, as I have posted here, my geometry is faked for earth and
moon as the separation is 1/10th real. Real distances and screen pixel
pixels just don't mix. That may also explain why NASA has produced so
few pictures of the two in the same image. 

	http://giwersworld.org/artiii/elevat23.jpg is an image with the correct
distance and diameters. (Don't hold me to anything else in it. I was
still learning to spell povray back then.) Not the white blob that in
fact is an image map of a moon photo. 

> I will try to explain it as clearly as I can and you can tell me where I go
> wrong.
> 
> Lets define the earth's orbit as a plane.  The sun is also on this plane,
> and in 99% of the illustrations of the system the camera is also in the same
> plane.  So far so flat.

	And the orbital plane is within a degree or so of the sun's equatorial
plane. I have to look that up also. 

> The earth's spin axis (through the north and south poles) is not
> perpendicular to this plane.  It is tilted by 23.5 degrees.

	And that is the tilt I dropped in for the image. 

> The moon is also not in the solar plane, but I will take your word for the
> angle.

	That is memory. I haven't come across the number in a long time. 

> So I guess what a photo would show is a vertical terminator that does not go
> through the north and south poles.  I hope that makes sense.  I have
> attached a tiny bitmap to illustrate the geometry.
> 
> In June everything north of the arctic circle is closer to the sun than the
> day/night terminator, which is why the sun doesn't set in the arctic in the
> summer.
> 
> In the December you get the exact opposite effect so the antarctic is in
> perpetual daylight and the arctic has perpetual night.  Spring and fall are
> obvious interpolations.
> 
> Enough geometry, on to POV :-)
> 
> If you model the earth as a single sphere with a single texture and light it
> from the "sun" you should be able to produce still shots and animations
> easily.
> 
> If you choose to use the "two textures" method things will be much trickier.
> You might be able to produce a snapshot by careful splicing of the day/night
> textures on the sphere, but a one-day animation (spinning earth) would
> require a geometry genius.

	C'est Moi! It might still be on p.b.a but vertical. 

earth-moon-sun-stars = union {earth + moon} + sun + stars

earth-moon = union {tilt}

===

	There should also be several stills here of 

	sphere {day_image} + sphere {night_image translate 0.0001*x}

> I hope you will take all of this in the spirit it is intended.  I really
> like your picture and I want to encourage you to produce one that combines
> the artistic merit you already show with an accurate geometry.  I wish I
> could do so well.

	If it is ever perfect I'll be the first to say so. ;) 

> "Matt Giwer" <jul### [at] ijnet> wrote in message
> news:3A3D788B.EF61EA76@ij.net...
> > You get it tilted.

-- 
I am never effected by my environment or anything else
for that matter. 
	-- The Iron Webmaster, 125

Post a reply to this message