> The image was rendered within one week

   Wow! ...it's a shame I don't have the patience to use my own includes for 
such long projects. ;)

   Very good work with the rainbows, both Mike and Florian. I've to agree 
with Florian that his one looks even more realistic due to the softer colors.

   Regards,

--
Jaime

Post a reply to this message

Jaime Vives Piqueres <jai### [at] ignoranciaorg> wrote:

> > The image was rendered within one week
>
>    Wow! ...it's a shame I don't have the patience to use my own includes for
> such long projects. ;)
>
>    Very good work with the rainbows, both Mike and Florian. I've to agree
> with Florian that his one looks even more realistic due to the softer colors.

I have to admit that there's still a little problem with my rainbows. In nature
you hardly can see ones with such soft colors; They indeed are a bit brighter.
Especially the blue band appears too blurred to me. This is because I wrote
some functions by myself to approximate the CIE visible light spectrum. I've
also tried the original values without approximation, but there I did something
wrong, the rainbow showed up 'out of color'. Instead of white, it was ugly
yellowish :)
But one day I will sit down in front of my PC and will begin to write a
brandnew, much better code for the color spectrum.

Greetings, Flo

Post a reply to this message

You've provided some very nice examples of what your include file can do.  I 
agree that my colors are too bright. I think the problem with the angle of 
the 'bows' is how I am determining the viewing and lighting angles. The 
project has been abandoned for now but if I do revisit this I may take a 
closer look at your implementation. The colors I used were just some values 
I hard coded into the source.

Regarding the render times, it was the isosurface that slowed things down 
along with two media containers; the one that comes with hf2iso and a box in 
front of the camera that I included to simulate scattered rain. The 
combination of several densities, scattering media, and isosurfaces really 
slows things down, especially since for each intersection with the media it 
has to check if any part of the isosurface is blocking the light.

I think that if you implemented your method in the emitting media in the 
source it would speed up render times several fold.

-Mike


"Florian Siegmund" <flo### [at] gmxat> wrote in message 
news:web.4952d7fd619c0ff387129fc00@news.povray.org...
> Take a glance at this one...
> The image was rendered within one week on a Pentium DualCore 3.6Ghz with 
> 2GB
> RAM; I don't know the exact time it took to render though, because I 
> stopped
> and continued tracing several times. For this image, I used Jaime Vives
> Piqueres' 'Project Tierra' landscape include files and my own ones for the
> rainbow, and I'm quite happy with the result. Still I'm using emitting 
> media
> rather than scattering one, because it renders faster and my include file
> contains the possibility to 'project' a pattern onto the rainbow, e.g. 
> clouds,
> to get the illusion of a 'broken' rainbow (not used in this image, but
> possible, if someone liked to use this feature).
>
> Mike, your rainbow seems a little bit too 'perfect' to me, like it was an
> illustration taken from a fairy tale book; most of all because of the 
> powerful
> colors. Maybe this is your aim, because the viewer still gets a very 
> special
> impression in this case, I don't know. If you wanted to achieve that, you 
> are
> going into the right direction (yes, I like the image; put a crystal 
> dragon
> into the scene and I would like it even more!) But for getting an image 
> with a
> more realistic look, you should work on the colors and angles of the 
> rainbow
> once more. One more thing to say, I wonder about your rendering times... 
> what
> is it that kills your processor speed, the rainbow media or the landscape
> isosurface? By default, my rainbow media only takes 2 media intervals. 
> There is
> no need of taking more intervals in normal cases, beacause I tweaked the 
> density
> function in a way so that it renders nearly like the inbuilt 'fog' feature 
> in
> pov-ray (assuming that reflecting water droplets are distributed 
> constantly
> along the viewing ray). Hope this helps a little bit.
>

Post a reply to this message

"Mike Williams" <nos### [at] econymdemoncouk> schreef in bericht 
news:Qs8### [at] econymdemoncouk...
> Wasn't it Thomas de Groot who wrote:
>
>>One thing I wonder about is the difference of luminosity of the landscape
>>inside and outside of the rainbow. I am not sure if this is a RL feature.
>
> Yes. The lightness of the rainbow itself has to come from somewhere. Some 
> of the light that would have come from the dark "Alexander's band" as you 
> see the rainbow is deflected to make the bright coloured stripes of 
> rainbows that are seen by people standing beside you.
>
> http://en.wikipedia.org/wiki/Alexander%27s_band
>
> -- 

Thank you indeed Mike, excellent explanation. I understand it now.

Happy Christmas.

Thomas

Post a reply to this message

Speaking of rainbows, I spotted this natural render yesterday.  Unfortunately,
my cheap digital camera wasn't up for the splendor. :P

My father told it was a sure bet to find a gold pot at its end, which looked to
be in reach for the people in those houses in the far right, a place
coincidentally named "Eldorado"... :D

Post a reply to this message

Attachments:
Download 'arco-iris.jpg' (555 KB)

Preview of image 'arco-iris.jpg'

Florian Siegmund wrote:

How do you make the "overall" (I don't see it in the sky) noise in both
of your pics? It makes them look like they were taken with a digital
camera in just a little bit too dark to be clean.

-Aero

Post a reply to this message

High!

Mike Hough schrieb:
> Well, I meant too long considering it was a test scene. I started out with 
> POV-Ray for DOS on a 166mhz system, so I appreciate the processing power 
> available these days.

I even started with Cyrix 80486 at 40 MHz... but back then, hardly any 
CPU needed cooling fans, so the computer was quiet enough to sleep 
beside it while rendering anything more complicated than a RSOCP (which, 
by the way, I NEVER did during my early POV-Ray days in the 1990s!)!

> I did this final render of this scene and wasn't going to post it because I 
> wasn't happy with it. It took 6d 9h 27m to render at 1920x1200.

Awesome! The mountains strongly remind me of the Hoggar mountains in 
southern Algeria - where rainbows are extremely rare!

See you in Khyberspace!

Yadgar

Now playing: She Knew Better (Munich)

Post a reply to this message

"nemesis" <nam### [at] gmailcom> schreef in bericht 
news:web.4953f734619c0ff3180057960@news.povray.org...
> Speaking of rainbows, I spotted this natural render yesterday. 
> Unfortunately,
> my cheap digital camera wasn't up for the splendor. :P

And Natural POV renders so much faster :-)
This is a nice example by the way.

>
> My father told it was a sure bet to find a gold pot at its end, which 
> looked to
> be in reach for the people in those houses in the far right, a place
> coincidentally named "Eldorado"... :D

Oooh, the lucky guys!


Thomas

Post a reply to this message

Among other things, Florian Siegmund saw fit to write:

> Another scene, another rainbow...

That's grat!

I seem to remember that, in addition to the two main full-spectrum bands,
there are some green-red stripes at one end... Do you simulate this too?

-- 
light_source{9+9*x,1}camera{orthographic look_at(1-y)/4angle 30location
9/4-z*4}light_source{-9*z,1}union{box{.9-z.1+x clipped_by{plane{2+y-4*x
0}}}box{z-y-.1.1+z}box{-.1.1+x}box{.1z-.1}pigment{rgb<.8.2,1>}}//Jellby

Post a reply to this message

I think what you are seeing is the noise component of the isosurface.


"Eero Ahonen" <aer### [at] removethiszbxtnetinvalid> wrote in message 
news:49540108@news.povray.org...
> Florian Siegmund wrote:
>
> How do you make the "overall" (I don't see it in the sky) noise in both
> of your pics? It makes them look like they were taken with a digital
> camera in just a little bit too dark to be clean.
>
> -Aero

Post a reply to this message