A Refractive Sphere on a Checkered Plain. (BTW, Archpawn, I *really* 
wish I'd come up with that pun myself.)

The photons notwithstanding ;-) this is my test bed for a cheap 
radiosity substitute.

Ambient wasn't doing it for me, since it is completely omnidirectional. 
I want surfaces facing upward (the ground) to get more light from the 
sky than vertical surfaces (tree trunks). Without taking this into 
account, the tree trunks tend to look like they glow in the dark (look 
at the tiles that Nicolas posted from his super render, and you can see 
what I'm talking about).

I'm playing with a custom lighting system derived from Tek's light dome. 
Normally, the light dome is significantly slower than radiosity. I tried 
hacking some of the quality optimizations out of his code and 
dramatically decreasing the light count (from 30+ to 5) without making 
much of a dent.

The secret is to make all the lights in the dome shadowless. My original 
IRTC render took ~3 hours; this image, at the same resolution and AA 
settings, took just over thirty minutes to render. Not too shabby. :-)

-- 
William Tracy
afi### [at] gmailcom -- wtr### [at] calpolyedu

    "I don't hink it helps to make analagies between the physical and 
social worlds," Sax said primly.
    "Shut up, Sax. Go back to your virtual reality."
     -- Kim Stanley Robinson, _Red Mars_

Post a reply to this message

Attachments:
Download 'refractive.jpg' (214 KB)

Preview of image 'refractive.jpg'

... and a giant version inspired by Nicolas. The jpeg compression is a 
tad icky, because I cranked the file to just under 500K in size.

I didn't have the light dome together when I rendered this; the ambient 
is turned off and a single shadowless light from directly overhead is 
filling in the shadows. Even that looks better to me than the ambient 
light. :-)

-- 
William Tracy
afi### [at] gmailcom -- wtr### [at] calpolyedu

These may involve hunting down whoever added whichever thing it is and 
torturing information out of them.
     -- from the GCC Documentation Project's web page

Post a reply to this message

Attachments:
Download 'refractive-big.jpg' (489 KB)

Preview of image 'refractive-big.jpg'

Looks good!
Could Jaime's Lightsys IV be an answer? I do not understand it fully but it 
gives good results.

Thomas

Post a reply to this message

Thomas de Groot wrote:
> Looks good!
> Could Jaime's Lightsys IV be an answer? I do not understand it fully but it 
> gives good results.

Lightsys looks a lot more elaborate than what I want. It seems to assume 
that you are already using radiosity, and want your light sources 
tweaked for maximum realism.

I just want something that renders fast without looking yucky. :-)

-- 
William Tracy
afi### [at] gmailcom -- wtr### [at] calpolyedu

Live by the code, die by the code.

Post a reply to this message

Looking good!

You're absolutely right; radiosity just isn't necessary as often as we use it.
With outdoor scenes like this I will sometimes cut it down to three small
lights --- the sun (direct light and shadows), a blue sky-light (for scattered
light from the atmosphere, no shadow) and a light of the predominant color of
the scene (as a super-rough approximation of the first-bounce of radiosity,
again no shadow). If I want a little more realism, I make 'em small area
lights.

-S

Post a reply to this message

I don't want to be below this giant crystal sphere...

Sven   :-)



"William Tracy" <wtr### [at] calpolyedu> schrieb im Newsbeitrag 
news:47b4f08a@news.povray.org...
>A Refractive Sphere on a Checkered Plain. (BTW, Archpawn, I *really*
> wish I'd come up with that pun myself.)
>
> The photons notwithstanding ;-) this is my test bed for a cheap
> radiosity substitute.
>
> Ambient wasn't doing it for me, since it is completely omnidirectional.
> I want surfaces facing upward (the ground) to get more light from the
> sky than vertical surfaces (tree trunks). Without taking this into
> account, the tree trunks tend to look like they glow in the dark (look
> at the tiles that Nicolas posted from his super render, and you can see
> what I'm talking about).
>
> I'm playing with a custom lighting system derived from Tek's light dome.
> Normally, the light dome is significantly slower than radiosity. I tried
> hacking some of the quality optimizations out of his code and
> dramatically decreasing the light count (from 30+ to 5) without making
> much of a dent.
>
> The secret is to make all the lights in the dome shadowless. My original
> IRTC render took ~3 hours; this image, at the same resolution and AA
> settings, took just over thirty minutes to render. Not too shabby. :-)
>
> -- 
> William Tracy
> afi### [at] gmailcom -- wtr### [at] calpolyedu
>
>    "I don't hink it helps to make analagies between the physical and
> social worlds," Sax said primly.
>    "Shut up, Sax. Go back to your virtual reality."
>     -- Kim Stanley Robinson, _Red Mars_
>

Post a reply to this message

Sven Littkowski wrote:
> I don't want to be below this giant crystal sphere...

Actually, I've been waiting for the "only you can prevent forest fires" 
jokes to start coming in. :-)

-- 
William Tracy
afi### [at] gmailcom -- wtr### [at] calpolyedu

In Fig. 3.18 we define two local variables, four and twenty (no jokes 
about blackbirds, please).
     -- Jeffrey D. Ullman, _Elements of ML Programming_

Post a reply to this message