Forum: Poser - OFFICIAL


Subject: BBLightMeter and Studio Formulas

GeneralNutt opened this issue on May 19, 2010 · 5 posts


bagginsbill posted Wed, 19 May 2010 at 4:13 PM

Quote - 1 2 4 is pretty much the standard portrait setting.  1 is the key light, the brightest.  2 is the fill and should read one f/stop dimmer.  4, the shadow fill, should be three f/stops dimmer.

That isn't an accurate statement, if the numbers 1 2 4 represent inverse ratios.

If you're saying the key light luminance is 1, and the fill light luminance is 2 (i.e. 1/2), then the fill light is indeed one stop darker.

But 4 is two stops from 1, not three.

Anyway... let's take a simpler case of just two white lights - an infinite, and an IBL. Suppose you want the brightest part of the surfaces lit by both to be 4 times as bright as the parts where the infinite does not reach. (Assuming the various surfaces have the same reflectivity.)

So you have some that are lit by Inf+IBL, and others that are only lit by the IBL.

You seek:

(Inf+IBL) / IBL = 4

Simple algebra says:

Inf+IBL = 4 * IBL
Inf = 3 * IBL

So - it's simple. Set your infinite light intensity to 3 times that of the IBL. The resulting ratio will always be 4 to 1.

Of course, we're talking about how much light falls on these surfaces, not how much is reflected. How much reflected light you see in the render will depend on the shaders (colors and the math of the various nodes.)

If you're trying to set up a three-light 1:2:4 ratio, things get trickier. If the fill light does not overlap the key light, then it's pretty easy. The fill light should be equal in brightness to the ambient light.

If the key and fill  do overlap, especially if it's a partial overlap as is usually the case, then it's tricky math.

Depending on the overlap, there are situations where the 1:2:4 ratio is simply not possible. This is because an infinite light alone produces a ratio of 1:2 over a curved surface if we compare the incident light at 0 degrees versus 60 degrees off-center. Any additional overlapping light will cause the 60-degree surface to be brighter than 1/2 of the 0 degree surface. So then we're talking about the brightest illumination being somewhere in between pointing at the key light and the fill light. The math is nasty.

I suppose some sort of meter could help find a solution. But I don't have time to work on it right now. I have a work deadline to meet and am going on vacation Friday for a week.

 


Renderosity forum reply notifications are wonky. If I read a follow-up in a thread, but I don't myself reply, then notifications no longer happen AT ALL on that thread. So if I seem to be ignoring a question, that's why. (Updated September 23, 2019)