Hello;

I hope that some of you seasoned users of poser can help me with this:

I make use alot of BB env sphere for my renders...only most of the time i just use the sphere alone which by itself has sufficient light for the whole scene...without adding any additional lights.
The problem is that if i add for example an indirect light to the scene, i find that it kind of washes out the scene, making it look to bright...then i try lowering the intensity of the indirect light, which helps a little bit...but i still dont get satisfactory results in my final renders.  And that is the reason that most of the time i only use BB sphere on its own, without any additional lights. Only one of the problems i get if i add no additional lights to the sphere, is that i get no shadows.

To summarize all this...i just would like some advice on what is the best light setup to use along with BB env sphere?...indirect light, point light, IBL lights, spot lights...distants?...
                                                                     
                                                                      
                                                                 
                          

Has any body used BB env sphere along with other lights, and get satisfactory results?...

I would really appreciate some advice on this matter.

Thanks for viewing, and God bless.

First thing to do is get a handle on light levels. As soon as you start mucking with combinations, you start to lose track of what the overall levels are and it's easy to overdo the lighting.

So - I'd suggest getting my light meter and rendering simple scenes with various amounts of this and that and see what the light meter has to say about what conditions you created. If you see red in it, you are overlighting.

Second suggestion - indirect light in Poser is too amplified to be in balance with how we set up speculars. Try going into the D3D render firefly dialog (it's in the Python scripts menu under partners) and reduce the IDL intensity to less than half. I don't know what the optimal value is, but 1.0 is way too high. When you have surfaces close to each other (such as an armpit) you see the amplification as an unnatural glow in crevices. When you don't have such surfaces, you still have the possibility of too much light and you feel like it is washed out but can't put your finger on it. The light meter will clarify this situation in an unmistakable way.

As for what lights to use with the env sphere, that really depends on the world you're creating. If it's a bathroom with a small window, the env sphere will contribute very little and you should be using a point light for every virtual light bulb in the room. If the mirror has 4 60 watt light bulbs over it, for example, you should be using four point lights at low intensity - something like 25 to 50%.

For outdoor sun light, a single infinite light is almost always the most accurate way to light it. But the intensity and color still vary based on situation. On an overcast day, the infinite should be muted (70%?) and have a wide shadow because the clouds are diffusing the sun's light. If it's beach day in the bahamas, I might set the sun to 130% and shadow blur radius should be exactly .5.

Bagginsbill I'm curious, do you set a light to 100 or more intensity and just control it's strength with the diffuse & specular swatch? Or do you just control the light with the intensity value and leave the specular and diffuse at white (unless you're turning one off completely or something)?

Like steering a car with your feet (possible but may produce unexpected results) it is possible to use gray chips for controlling light intensity, but I use the intensity parameter to control intensity and leave my chips white.

Color chips that are not white or black are altered by gamma setting. If you do not firmly understand what RGB 170, 170, 170 really means with and without GC, and you're experimenting with gamma as an artistic control, you may be changing more than you mean to.

The concept of "equivalence" in renderer lighting does come up occasionally and I see people arguing that you should do color chip variation and keep intensity 100.

But let's examine this notion. (Sticking with gray scale color chips for the sake of simplicity.)

The effective strength of a light is the color multiplied with the intensity:

C * I

Where C means the color chip value and I means the intensity.

However, since color chips are affected by gamma, that's not the whole truth. The whole truth is:

(C ** g) * I

Given a light with a color chip set to RGB 186, 186, 186 and intensity 80%, what is the effective strength?

((186 / 255) ** 2.2) * .8 = 40%

Did you expect the light to be so dark when using a light gray? Probably not.

The key point: color values are nonlinear. Making a modest decrease in chip color will produce a dramatic decrease in brightness. This is intentional. It let's us represent a wide range of colors using only 8 bits. That's why sRGB is set up that way. We need lots of dynamic range. Linear values would not give us enough range to work with. The non-linear nature of digital color values is helpful in achieving realistic looking images with fewer artifacts - linear values would look banded as you got down to the lower values.

Second key point - humans are not good at thinking nonlinear.

So - you're asking for trouble if you mess with a nonlinear control system using your linear intuition.

For my case above, an equivalent setting is WHITE color chip (C = 1) and I = .4.

((255 / 255) ** 2.2) * .4 = .4

Now if I want to increase intensity by just 10%, I can type in 44%.

If you can figure out how to get 44% using gray, then by all means you can use the equivalence of factors in multiplication, but you have to take into account that the C value is nonlinear. If you can do that, use C when you feel like it. If you can't, use I instead.

Quote - Like steering a car with your feet (possible but may produce unexpected results) it is possible to use gray chips for controlling light intensity, but I use the intensity parameter to control intensity and leave my chips white.

Color chips that are not white or black are altered by gamma setting. If you do not firmly understand what RGB 170, 170, 170 really means with and without GC, and you're experimenting with gamma as an artistic control, you may be changing more than you mean to.

Yes, this is what I'm trying to avoid, experimentation without fully understanding what is going on.

However, ArtBee wrote a Gamma Correction guide:

http://www.book.artbeeweb.nl/?p=317 - PDF on the page.

On page 31 he talks about Anti Gamma Correction and mentions this:

"By putting the light intensity in the color swatch instead of in the numericals, you do get the better image. This holds in all cases of the Material Room."

He compares one image with 100% white and 60% intensity vs another image with 100% intensity and 60% white. I'm still trying to make sense of it. 

I use .7 IDL intensity. I have a single point light @ 112% intensity in a GC-2.2, IDL, SSS scene, enclosed room. The point light specular is at 100% and diffuse set to 244,244,244. The render looks great I don't see any problems or washing out. The cool thing about this lighting strategy is I don't have to add another light just for specular. I use your light meter and both the diffuse & specular are as close to white as possible without yellow or red burning.

Quote - He compares one image with 100% white and 60% intensity vs another image with 100% intensity and 60% white. I'm still trying to make sense of it.

This is exactly addressed in my subsequent post. These are not equivalent. Attempting to treat them as equivalent leads to misunderstanding.

I do not mean to put down what artbee wrote, but if he wanted to make the point accurately, the equation involves gamma - it's not a straight multiplication.

The notional difference between C=100 and I=60 versus C=60 and I=100 comes due to misapplication of equivalence. They are not equivalent, but they can be made equivalent if you use the correct math.

To get equivalence for 60% using C, you want to plug in to the equation and solve for C.

(C ** 2.2) * 1 = .6

C ** 2.2 = .6

C = .6 ** (1/2.2)

C = .79

C = RGB 202, 202, 202

Thus we see that RGB 202 and I=100 is equivalent to RGB 255 and I=60

That is not what you were told to compare. Had you compared what I just wrote instead, you would find zero difference - they are exactly equivalent.

You posted just before I did lol! Thanks for this explanation. I'll be meditating on it for a while over coffee.

Question, BB...

Do you ever recommend lowering the light level from the Env Dome to compensate for brighter point lights? If so, where is the best node to adjust the level in the Env Dome.

Quote - Question, BB...

Do you ever recommend lowering the light level from the Env Dome to compensate for brighter point lights? If so, where is the best node to adjust the level in the Env Dome.

That's a really good question, I'd like to know as well!

If you are using the panoramic shader on BB EnvSphere, locate the HSV node and decrease the Value in order to decrease the brightness of the Sphere or if the sphere is not visible in your scene, use a darker image.

If you use the gradient material as the background image, insert an HSV node and perform the same as Latexluv suggests.

Never have tried the gradient material. I should try it sometime.

I find the biggest headache when mixing lights is self lighting materials, catches me out a lot. Have started using a set workflow going from lowest intensity to highest. Its dumb and takes more time but I find it is less frustrating.

Very informative post and discussion, I have been trying to add colour to shadow and have clearly been using the wrong approach. Thanks

Coloring shadows is definitely the wrong approach unless you are trying to achieve a surrealistic scene.  IBL and IDL will 'tint' your shadows because of ambient lighting.  Self illuminating material typically is caused by either an active ambient channel or light emitting checked in the Properties panel.  In P9/PP2012, the diffuse_value attribute has recently been discovered to be out of whack with IDL; you need to reduce the value extremely low and use higher light intensities, or reduce the IDL intensity in D3D's Render settings to less than one (0.15 - 0.3) to achieve the same affect with no self illumination where mesh makes contact.

Yes, the effect I was trying to achieve is caused by materials reflecting the blue in the sky and not from absorption, reading this thread made me realise it was nonsense trying to use light blue in the shadows using a light blue ambient light.  Thanks

Thanks Sir Ted!

I've been using your Sphere for most renders-and was just using IDL at 100% and a specular only light. I'll try IDL at 0.5 and an infinite light instead.

I wonder if any of this including reflections will be changed in any way with PoserProX (2014)?

sorry if somebody already mentioned, however use bill's envsphere with smart IBL method, in which parametric location of sun on equirectangular hdri in (u,v) co-ords is used to position sun (poser inf lite) to correspond to sun location on associated hdri, using simple cartesian/polar co-ordinate conversion math.  see hdrlabs.com for explanation of sIBL method.

in addition, OP is advised to quit using IBL/AO until somebody explains how to use that (poser 7 and earlier).