Hi, I wrote several even-distribution light source macros. The technique to
place any number points on sphere is what Samuel Benge reply to me several
months ago.
And I spent much time to research how to decide light's intensity. I welcome
your advice.

Post a reply to this message

Attachments:
Download 'idealarealight.inc.txt' (11 KB)

On 1/15/2014 10:30 AM, And wrote:
> Hi, I wrote several even-distribution light source macros. The technique to
> place any number points on sphere is what Samuel Benge reply to me several
> months ago.
> And I spent much time to research how to decide light's intensity. I welcome
> your advice.
>

Does your code integrate well with LightSysIV? I would like softer 
shadows, but not if it messes with the light intensity.

-- 
http://isometricland.net

---
This email is free from viruses and malware because avast! Antivirus protection is
active.
http://www.avast.com

Post a reply to this message

posfan12 <pos### [at] gmailcom> wrote:
> Does your code integrate well with LightSysIV? I would like softer
> shadows, but not if it messes with the light intensity.

Mmmm....Not yet.
So, it is better to integrate it with a existing package. It is good to hear
some feedback. I'd like to prove and increase my ability in coding technique.
I'm not familiar with LightSys yet then try it later. But I explain my code
first.

This include file use physical quantity like nit(cd/m^2) and lumen. The global
parameter NITS_PER_WHITE can adjust all brightness where my macros are used. The
eye or other photometry device is directly sensitive with luminance so in a
rendered image the pixel's brightness can directly associate with luminance.(But
first the assumed_gamma should be 1). Luminance's unit is nit=candela/m^2.
Here is some luminance value in wikipedia:

https://en.wikipedia.org/wiki/Orders_of_magnitude_(luminance)

I usually set NITS_PER_WHITE = 200~1000 in a indoor scene when it has several
light sources and each one have flux = 1500~3000 (lumen).
For a outdoor scene with sun, NITS_PER_WHITE should increase to 2000~10000.
The NITS_PER_WHITE = 1000 means that when a pixel's luminance=1000 nits then it
shows a <1,1,1> white in rendered image, where the luminance=500 nits point it
show <0.5,0.5,0.5> gray.

The macro still have big limitation in that it only deal with gray scale's
intensity. which I may take a research on LightSys(I know that deal with color),
the other big limitation is that it assume the POV-Ray unit is 1m. I will add a
global parameter MM_PER_UNIT to adjust that later.

Post a reply to this message