This has one emitter as well.  6 Ray bounces. The rest was pretty draft. The specular and bump maps are matched.

Interesting discussion. Just a question:

What mapping do you people use? I found in previous experiments with (sun) lights set to values in the order of 10 to get some light in the corners of a room lit from outside that exponential HSV can make quite a difference in 'taming' overly dark shadows and washed-out lights.

Hi,

Yes. HSV value upping the dome light to 8X or 800% depending on the HDR and scene, etc. But, Reducing IDL intensity in D3D rendersettings from 1.0 to as little as 0.1 to try to balance diffuse anomolies in Poser. That's how I understand it. I know something is almost always out of whack.

Quote - Hi,

 

Yes. HSV value upping the dome light to 8X or 800% depending on the HDR and scene, etc. But, Reducing IDL intensity in D3D rendersettings from 1.0 to as little as 0.1 to try to balance diffuse anomolies in Poser. That's how I understand it. I know something is almost always out of whack.

Do you ever use more that one directional light indoors or outdoors?

Do you ever use more that one directional light indoors or outdoors?

Nice hook. Sure.  You can have 5 suns parented together pitched at slightly different angles or 50 point lights in a candleset... They cost you in rendertime, though. This above is set up to test some ideas based on bb's idea. I have more than one light. There is an emitter and the sphere

Doing what BB mentioned with the IDL intensity works extrememly well in indoor scenes that have a lot of shadows from external light.

I was getting the dreaded red bleed on this one, and it is gone now that I turned down the IDL intensity to 0.1 and turned up the emitters in the scene to add more light.

 (Bigger version in my gallery)

Everything in the scene is visible in raytracing and set to cast shadows. I have not found a need to hide the hair, or anything else for that matter. Ezskin is applied to the hair and the body, using the sweaty effect on the skin. I did have to turn down the eye relfections, they were so reflective it hid the iris texture. Cant really see it in this render, but I noticed it in a few test renders I did from different angles.

I am still experimenting to see what other things need to be changed when doing this. I did notice that there can be a color shift when doing this, but it is not that drastic. There is also a small difference in the shadow edges based on what the radius blur is set at. I had to turn the shadow samples way up to compensate.

Doing all of this does raise the render time a tad, but the results are more than worth the wait.

Doing all of this does raise the render time a tad, but the results are more than worth the wait.

I'd have to agree. Why no reflect on anything? Hust to speed up the test?

Quote - Doing what BB mentioned with the IDL intensity works extrememly well in indoor scenes that have a lot of shadows from external light.

I was getting the dreaded red bleed on this one, and it is gone now that I turned down the IDL intensity to 0.1 and turned up the emitters in the scene to add more light.

 

 (Bigger version in my gallery)

Everything in the scene is visible in raytracing and set to cast shadows. I have not found a need to hide the hair, or anything else for that matter. Ezskin is applied to the hair and the body, using the sweaty effect on the skin. I did have to turn down the eye relfections, they were so reflective it hid the iris texture. Cant really see it in this render, but I noticed it in a few test renders I did from different angles.

I am still experimenting to see what other things need to be changed when doing this. I did notice that there can be a color shift when doing this, but it is not that drastic. There is also a small difference in the shadow edges based on what the radius blur is set at. I had to turn the shadow samples way up to compensate.

Doing all of this does raise the render time a tad, but the results are more than worth the wait.

That looks great, do you have any similar scenes where she is nude?

Quote - Do you ever use more that one directional light indoors or outdoors?

Nice hook. Sure.  You can have 5 suns parented together pitched at slightly different angles or 50 point lights in a candleset... They cost you in rendertime, though. This above is set up to test some ideas based on bb's idea. I have more than one light. There is an emitter and the sphere

At the most basic level, I think IDL with one directional light is complete garbage.  Does everyone agree? 

1. One envsphere with nothing on it set to white diffuse. (no nodes nothing just white)
2. One directional light pointing at figures & scene set to 100% strength
3. Raytrace Bounce 1
4. IDL Bounces 2
5. BB Light meter to show when there isn't specular or diffuse burn.
6. IDL intensity set to 1

The above does not work well from my experiments.  Maybe I'm supposed to turn on gamma correction? I'm trying to get IDL to look good at the most BASIC of levels.

There is reflect on just about everything, but they are turned way down because the ceiling of the room is the main emiiter. I need to redo it with a plane and hide it from the camera.

Basically you hit the nail on the head, this was more of a test then anything else.

BB is definently onto something here. If I posted the before version with all the red bleed around the shadows on her from the blinds you would see a drastic difference.

They looked awful, lol...

Just to clarify the lighting, there are 2 lights in this scene. a spot light just outside the windows, and a point light in the floor lamp. The rest of the light is emitted from the cieling, both bounces are set at 2.

I will work with it a little more and see what I come up with. Looks like we have to rethink Poser lighting again. The results are getting to the point that using other render engines may not be nessasary at all to achieve stunning results.

Quote - That looks great, do you have any similar scenes where she is nude?

Not yet, but I wanted a lot in the scene to see how it all worked together.

Quote - At the most basic level, I think IDL with one directional light is complete garbage.  Does everyone agree?

It depends on the scene, and how complex the emitter setup is. I rarely use the enviroment sphere now. Don't get me wrong you can get stunning results with it, just not so much in a room.

If I turned off the floor lamp in that scene, which is just a point light, there would only be one light left.

There is obviously more that we can learn about how to set up IDL to get the best results, we just have to mess around with it and find what that is.

Quote - There is obviously more that we can learn about how to set up IDL to get the best results, we just have to mess around with it and find what that is.

That's the problem, I don't want to mess around. I want to get some work done.  All this experimenting is getting old. I want a solid workflow & methodology so I don't have to worry about this crap and just focus on content creation.

Smith Micro is fail.

I expect IDL at the most basic of levels to just look great, then from there if I want to tweak stuff so be it.

Quote - Just to clarify the lighting, there are 2 lights in this scene. a spot light just outside the windows, and a point light in the floor lamp. The rest of the light is emitted from the cieling, both bounces are set at 2.

See that is confusing. I would expect a directional light instead of a spot light no? You've achieved great results but the methodology is unorthodox isn't it? Letting a ceiling light your scene?  Why can't it be simpler?  Having to go through each item in your scene is too time consuming.

ust to clarify the lighting, there are 2 lights in this scene. a spot light just outside the windows, and a point light in the floor lamp. The rest of the light is emitted from the cieling, both bounces are set at 2.

Did you notice any artifacting issues? I was wondering if raising emitters and lowering IDL intensity is really increasing Ray success. I mean, it's missing the target less and not leaving blanks/splotchies.

Cannot wait to post my next render... sheeez

The spheres I was using was for exterior shots but I tried to bounce it into the room with 6 or 7 bounces. It's just easier to do it with a spot We still need a way to emit light and still be able to be selected invisible to some other selective surface. This was what carodan was after, I think. Reflections in the eyes are a perfect example when using hot emitters.

Why does BBEnvsphere make such a huge difference? Is IDL in poser pro broken by default?

Quote -  The spheres I was using was for exterior shots but I tried to bounce it into the room with 6 or 7 bounces. It's just easier to do it with a spot We still need a way to emit light and still be able to be selected invisible to some other selective surface. This was what carodan was after, I think. Reflections in the eyes are a perfect example when using hot emitters.

Yep, that's the problem. Reflections go out of whack with the powerful emitters.

There's such a variance in HDRi quality & setup too. I have some that almost don't require a shadow casting light at all, some that do but don't need anywhere near as much power as you'd think

I'm also starting to wonder about specular & fresnel reflection strength - find I'm really lowering those values a lot. Not being the scientific mind, I'm almost certainly getting a few things mixed up here & there.

I was wondering though how people were setting up their Environment Spheres or emitters. On bbs EnvSphere I've tried varying HSV values but also messed with the incoming gamma (using the nodes he included). Also gone for the raw HDRi straight out of the Ambient channel. All give ways of balancing out the final strength of the HDRi.

Hopefully bb will cast his wisdom among us again soon - great fun floundering about in his absence.

That render of mine is almost done... and there was blinking toe glow. He he.

Not to worry was planning to post-work the hell out of this one anyway.

About 5 more buckets of that render to go and then I can have a play myself with this new IDL lighting paradigm... :)

Quote - Smith Micro is fail.

Feel free to set your scenes up in Houdini, Blender, Max, or any other 3D app, and you will find that it takes just as long if not longer to set them up. You will have to set up network nodes out the wazzu in Houdini for both the scene, and the textures. If your using a render engine other than Mantra you will have to set the textures up for each additional engine. Blender can be set up for tons of render engines, that require different texture setups for each one. Max can use tons of render engines as well.

I don't think Smith Micro is fail at all. It does an excellent job for pennies on the dollar compared to other applications. Price out a Houdini or Max suite and you will quickly see what I mean.

(Houdini can easily hit 7 figures with just a small render farm. Queue manager that comes with Poser Pro supports 9999 simultanious nodes at no additional cost.)

Show me another application that comes with a full fledged render engine. Before you say Daz, keep in mind that the version of 3Delight it comes with is nerfed compared to the full version you can buy. Blender is the only free program that can hold a candle to Poser in money spent vs what it can do. Blender is free... Cycles (Blenders new render engine) is still in developement, and changes constantly. Scenes set up for earlier releases sometimes need adjustments to work with later versions.

If you want a setup that is click and go, good luck with that. Movie studios don't sell those to the public simply because it takes way to much cpu power to do much of anything in them once you click render.

Quote - See that is confusing. I would expect a directional light instead of a spot light no? You've achieved great results but the methodology is unorthodox isn't it? Letting a ceiling light your scene?  Why can't it be simpler?  Having to go through each item in your scene is too time consuming.

I used a spot light to simulate a light that was right outside of the window. I would use a infinite for daylight.

Quote - Feel free to set your scenes up in Houdini, Blender, Max, or any other 3D app, and you will find that it takes just as long if not longer to set them up.

That's the point, why do you think I'm using Poser Pro? I'm using a "LIGHT" 3d application to increase my productivity with LESS downtime.

i'm purposely avoiding applications like houdini, blender and max so I can avoid all the BS that I'm not interested in.

Houdini, Blender, Max, Maya = LOTS of downtime.

Poser Pro = Zero downtime before SSS & IDL was introduced.

Smith Micro is fooling themselves if they think that poser is going to compete with the big boys. Smith Micro needs to realize what Poser Pro is. A LIGHTWEIGHT 3D app for beginner to intermediate level users.

Smith Micro should have SSS & IDL solutions so people can get straight to producing content instead of messing around constantly.

Quote - I used a spot light to simulate a light that was right outside of the window. I would use a infinite for daylight.

Wouldn't you use an infinite at low strength for night light? You're the expert btw, I'm just trying to adhere to a standard that makes sense instead of doing shortcuts and unthordox methods, why? Because then I'd be spending far too much time in each scene.

Attached Link: http://www.renderosity.com/mod/gallery/index.php?image_id=2392902

I'm kinda blown away by that. Not sure yet what the settings did for it but I add them here to show it was no big deal,, samplewise. I wanted to do this in a nice planned way but got distracting results like link...;) I'm done polluting the thread now so, thanks and let's see what happens.

Quote - Smith Micro should have SSS & IDL solutions so people can get straight to producing content instead of messing around constantly.

If SM figured out how to make a one click solution to SSS and IDL it would put the big boys out of business. The big boys have not figured out how to do it.

There is no such thing as a one click fits all in anything 3D. Not from SM, Side Effects, Autodesk, nobody...If you use any of those apps you would know what I mean.

I have Houdini 12, and it doesn't have a one click solution for anything unless someone else did the ground work for it. There are a few nice presets available for it to get you real close, just like there are for Poser and other 3D apps. But they are far from perfect out of the box 99% of the time.

I have Blender, same deal. There are presets that will get you close, but thats about it.

I don't have a newer version of Max, but I doubt it has anything better than the next program when it comes to one click solutions.

Quote - Poser Pro = Zero downtime before SSS & IDL was introduced.

If you had no downtime without using SSS and IDL, then stop using it and go back to the way you did it before with zero downtime. All the stuff that was there before, is still there now.

The downtime you have now is self inflicted, by your own statement. If you can't get it to work, just don't use it. If you use it for a production enviroment I can understand why you would not want to waste time trying to figure it out. If you are a hobbists, you are like a lot of others that just want to get closer to a one click solution.

We will get closer, but a one click solution wont happen anytime soon.

Quote - > Quote - Smith Micro should have SSS & IDL solutions so people can get straight to producing content instead of messing around constantly.

 

If SM figured out how to make a one click solution to SSS and IDL it would put the big boys out of business. The big boys have not figured out how to do.

There is no such thing as a one click fits all in anything 3D. Not from SM, Side Effects, Autodesk, nobody...If you use any of those apps you would know what I mean.

I have Houdini 12, and it doesn't have a one click solution for anything unless someone else did the ground work for it. There are a few nice presets available for it to get you real close, just like there are for Poser and other 3D apps. But they are far from perfect out of the box 99% of the time.

I have Blender, same deal. There are presets that will get you close, but thats about it.

I don't have a newer version of Max, but I doubt it has anything better than the next program when it comes to one click solutions.

Quote - Poser Pro = Zero downtime before SSS & IDL was introduced.

If you had no downtime without using SSS and IDL, then stop using it and go back to the way you did it before with zero downtime. All the stuff that was there before, is still there now.

The downtime you have now is self inflicted, by your own statement. If you can't get it to work, just don't use it. If you use it for a production enviroment I can understand why you would not want to waste time trying to figure it out. If you are a hobbists, you are like a lot of others that just want to get closer to a one click solution.

We will get closer, but a one click solution wont happen anytime soon.

Wait, do you work for Smith Micro?

On a side note, I'm almost finished with a methodology that will work on a basic level for indoor & outdoor scenes.  So instead of dicking around rendering over & over again, I can reliably take 5-10 VERY VERY SIMPLE steps (starting from an empty scene) and crank out something nice to look at.

Here is what I expected from Smith Micro.

1. Empty Scene

2. Add Figures & Props

3. CLICK - SMITH MICRO'S IDL OUTDOOR SOLUTION - Click (Automatically adds lights & sphere for outdoor) and expects you to aim the primary light at your figure.

4. SMITH MICRO'S SSS SOLUTION FOR SKIN - Click (Applies SSS to skin with whatever settings you want)

5. Render

DONE with a basic, nice looking scene with IDL & SSS. From here I can tweak it if I want.

I almost have the above working but it took 4 frickin days which sucks.  If a complete noob like me can figure out a basic methodology think what the big boys at smith micro could do if they actually put the time in.

HERE IS WHAT I AM TRYING TO AVOID:

"The scene doesn't look right and I have no idea why. Let me do some realy weird out of the box BS.. render.. tweak.. render.. tweak.. render.. tweak... render tweak...."

2 days later

"MAN the scene finally looks great but what I did makes absolutely no sense!!!?!"

Two V4 based figures with Syyd Raven's Oxygen skin shaders and Zev0's Vascularity applied.

2 x Envsphere with the same texture I rendered in Vue - the Outer with HSV value boosted to 2.0, inner with Visible in Raytracing off.

1 x Infinite Light at 88%

Various Dreamland items, and Catbot, with Bagginsbill shaders as a further benchmark.

Rendered with IDL = 0.5

You didn't figure in that a good deal of content out there are incorrectly set up.  Even if your methodology works with your set up, would you be willing to test several different models and textures to validate your findings?

Bwah bwah mommy! This guitar sucks! I want to sound like God even if I play shit! The designer is a moron! Gibson and Fender - fail!!!

I don't know about anybody else, but I have always found such childish moaning to be entirely tedious.

On the other hand, I do enjoy discussions about how we might all improve our art.

My compliments...

Quote - You didn't figure in that a good deal of content out there are incorrectly set up.  Even if your methodology works with your set up, would you be willing to test several different models and textures to validate your findings?

In terms of my own methodology, I was planning to start with a benchmark of BB derived shaders... i.e. the Syyd Raven Oxygen skin sets I've started with use EZSkin shaders as their basis.

I'm currently doing another version of my test render, this time with IDL lowered to 1.5 and the IDL-illuminating envsphere amplified, in the HSV node value, to 6.6666667 (or thereabouts).

Infinite light the same, at 88%.

Once I see what that looks like, I will maybe try some other skins I have... fed through EZSkin2 probably - but with the defaults for EZSkin2 perhaps... as I expect Syyd has fairly tweaked those. So it would be good to see how the default EZSkin settings perform I guess??

Will maybe see how Blackhearted's Tyler, GND4, Anastasia and Shae textures perform too, I reckon...

EDIT: At the moment I'm going with the goal of not adjusting the shaders. Only IDL intensity and lighting...

He is looking for the same thing a lot of us are. The elusive Make Art button...

Quote - You didn't figure in that a good deal of content out there are incorrectly set up.  Even if your methodology works with your set up, would you be willing to test several different models and textures to validate your findings?

I have no choice, that's what I'll be doing for the next few days. I'll be converting all my my pre-SSS, pre-IDL scenes.  

Soon enough I'm going to make a thread that shows generic SSS & IDL lighting setups for outside (daytime), inside (daytime) & indoor (night). But I'm no expert, so I'd look forward to any help! :)

Quote - You didn't figure in that a good deal of content out there are incorrectly set up.  Even if your methodology works with your set up, would you be willing to test several different models and textures to validate your findings?

I have no choice, that's what I'll be doing for the next few days. I'll be converting all my my pre-SSS, pre-IDL scenes.  

Soon enough I'm going to make a thread that shows generic SSS & IDL lighting setups for outside (daytime), inside (daytime) & indoor (night). But I'm no expert, so I'd look forward to any help! :)

Quote - He is looking for the same thing a lot of us are. The elusive Make Art button...

1. Empty scene

2. Add IDL Lighting (Night)

3. Add BBenv

4. Add Props & scene, position.

5. Add V4 (Turn on bblight meters)

6. Pose

7. Move point light to optimum spot (there are other lights in the scene that I move into a default position no matter what the scene).

8. Click three times for SSS skin setup

9. Render

10. Review

11. 3-4 more renders and adjustments. Do any props need AO to ensure contact is made between objects?

12. Final Render

Once scene = 3-4 hours tops.

The above is always my goal, if I go outside of that it means I'm doing something wrong.  You have to give yourself limits.

Definitely going to be necessary to get into adjusting the reflectivity in shaders at these levels, I guess...?

Catbot's colour was changed to Pink Flamingo Candy Satin Flame just for the sake of confusing matters, by the way... ;-)

I don't recall having problems with glow in occulded areas in Poser 8 and PP2010 (of course, I wasn't using the shaders I am now).   Is is this new to P9/PP2012?

Quote - I don't recall having problems with glow in occulded areas in Poser 8 and PP2010 (of course, I wasn't using the shaders I am now).   Is is this new to P9/PP2012?

SSS is involved, as I understand it, with the glow thing... not just IDL.

MonkeyCloud, I turned the reflection on the corneas way down.

Poser Surface reflection value to .01, and Anisotropic Specular value to .01. I am sure I could go higher, but I still need to experiment with it.

If you do any of the FX in Ezskin, there is a bunch more that needs turned down on them as well.

I think that 4 is about the max emitter level that will work without washing things out and creating artifacts. If you need more IDL light adding more emitters looks better than turning one up and blowing out the calculations.

Keep in mind that any calculation that is out of range (0.0.0 - 255.255.255) is doomed from the start.

Which reminded me... I need my base benchmark. Here's my quickie scene with the defaults of IDL = 1.0 and Envsphere HSV value = 1.0...

...oh... hang on a minute... no glow. Hmmm, I need to start again with a test scene with some glow in it. He he.

Actually, personally, I think this version looks the best??? Even with my fairly lo quality settings here of shading rate = 0.7 and Irradiance cache = 20...

Quote - MonkeyCloud, I turned the reflection on the corneas way down.

Poser Surface reflection value to .01, and Anisotropic Specular value to .01. I am sure I could go higher, but I still need to experiment with it.

If you do any of the FX in Ezskin, there is a bunch more that needs turned down on them as well.

I think that 4 is about the max emitter level that will work without washing things out and creating artifacts. If you need more IDL light adding more emitters looks better than turning one up and blowing out the calculations.

Keep in mind that any calculation that is out of range (0.0.0 - 255.255.255) is doomed from the start.

Thanks Shvrdavid...

...not sure now even how good a test this outdoor scene is even. Maybe I need an indoor... or partly indoor scene...

But I might just try this one again with IDL Intensity = 0.25 and Envsphere value = 4, just for kicks... ;-)

The issue is to avoid blowing out the other elements in the scene too... compare the colours between that last baseline (IDL Intensity = 1.0) with the reduced IDL intensity renders and the colours definitely start to wash out... I think? I think even at IDL Intensity = 0.5? Or is it my eyes and the late hour here?

...not sure the emitter light sources should be exactly doubled, just because IDL Intensity is being halved... I am wondering if the mathematical relationship between these two isn't more complex perhaps???

It probably is a little more complex that simple math. I don't know the formulas used in some of this so figuring that all out would require a little digging on what they actually are.

This is promising thou, I just a few hours of experimenting I am getting very good results with 1 light for shadows, and 2 emitters set to 4 amb. IDL level is set to .1

I am up to 1.2 on the saturation of the textures as well. When this one is done rendering I will post it and the settings I used for it. I am using on of my textures with lots of tattoos to see if it works with them as well.

Ok, this is what I have come up with so far.

These are the render settings I used.

And this is what I got.

The hair needs still needs some work thou. Parts of it are blown out.

Scene has one infinite light set to casting shadows set to 66% intensity and 3 planes.  2 of the 3 are set to 4 ambient and are white for IDL lighting. The top plane is all but as big as the ceiling, right up against it, and the other one is about the size of the character slightly in front of her at an angle facing out the rear window. The plane with the background on it is set to 6 ambient to self light it.

Skin is set to 1.2 sat with standard Ezskin, but main bump is turned up to 0.01 (my units are in inches) Cornea reflections and anistropic specular are turned down to 0.01 as well

It took about 8 min to render.

Hm, something else seems to have changed too. It used to be that increasing IDL bounces greatly increased light spread, so that you didn't need extra intensity to lighten up dark areas. My initial testing indicates that may no longer be the case.

Anyone else checking in this area? Maybe I don't have a complex enough scene to make a different yet, I dunno....

Quote - Hm, something else seems to have changed too. It used to be that increasing IDL bounces greatly increased light spread, so that you didn't need extra intensity to lighten up dark areas. My initial testing indicates that may no longer be the case.

Anyone else checking in this area? Maybe I don't have a complex enough scene to make a different yet, I dunno....

I've wondered about this too.  If you use a single directional light, the edges of the figure are black.  This is why I always say one dir light is crap, but then other people seem to be making it work somehow?

I'm going to try an indoor test scene next I think.

Quote - Hm, something else seems to have changed too. It used to be that increasing IDL bounces greatly increased light spread, so that you didn't need extra intensity to lighten up dark areas. My initial testing indicates that may no longer be the case.

Anyone else checking in this area? Maybe I don't have a complex enough scene to make a different yet, I dunno....

It must be that having the IDL intensity to 1 is just as BB stated. It is way to much and adding bounces just compounded the problem, increasing the brightness to much with each bounce.

With the intensity turned down, the light still bounces, but it doesn't drastically increase the brightness with each bounce.

I real world lighting each bounce should only increase the brightness by a small amount off of most surfaces anyway.

Been lurking here with great interest. Now keep in mind I'm wrong a lot (really, you can ask my wife), how I interpreted Bagginsbill's post is that he didn't say anything about increasing ambient to get the brightness back. I read that you should just lower IDL strength to .1 or .15 and see how the render looks.

What I'm thinking is that he is saying that too much of the scene brightness is coming from IDL, and that we should be lowering IDL level, then tweaking the lights for the correct levels.

Doing this with render gamma at 2.2 seemed to result in marked improvements to the test scene, like an improved VSS render.

Anyway, that's my take on it.
Image at
http://www.renderosity.com/mod/gallery/index.php?image_id=2393107

bevans: increasing ambient in a scene with no lights is analogous to increasing light brightness. If it's a "no lights" scene, obviously tweaking the lights isn't an option. :)

Even if there are lights in your scene, if you're using objects as light sources, upping their ambient value would also be analogous.

So it really depends on the scene setup.

Quote - Been lurking here with great interest. Now keep in mind I'm wrong a lot (really, you can ask my wife), how I interpreted Bagginsbill's post is that he didn't say anything about increasing ambient to get the brightness back. I read that you should just lower IDL strength to .1 or .15 and see how the render looks.

What I'm thinking is that he is saying that too much of the scene brightness is coming from IDL, and that we should be lowering IDL level, then tweaking the lights for the correct levels.

Doing this with render gamma at 2.2 seemed to result in marked improvements to the test scene, like an improved VSS render.

Anyway, that's my take on it.
Image at
http://www.renderosity.com/mod/gallery/index.php?image_id=2393107

If you're reducing the IDL intensity and your light balance was otherwise good, when IDL intensity was set at 1.0, then you're surely going to need to add some more light, to rebalance it?

That will either mean increasing the Poser lights... effectively meaning you're just using less IDL, in the overall mix, and more Poser (Infinite, Spot, Point) light.

Or it means increasing the IDL lighting (which might be all the lighting you have in a scene anyway, if you're not using Poser lights).

But that might just be at the level of "tweaks" rather than any more dramatic increase...

Certainly, from my loosely structured experiments so far, it seems that if you divide your IDL of 1.0 by 6.666..., to take it down to 0.15... it isn't then just a case of multiplying the ambient lighting by the same, 6.666..., to rebalance the lighting... far from it.

The increase needed in the lighting, ambient or Poser lighting, to compensate for the reduced IDL, seems to be much smaller than the increment IDL has been reduced by...? Sound right...?

Of course this must be likely to vary, possibly substantially, depending on the scene... I guess.

Keep going. bevans84 was right in that I didn't say to increase the emitter lighting (such as the EnvSphere), since what I was most trying to address was the surface-to-surface bounced light that was resulting in light amplification in crevices. I'm trying to tone that down since it is clearly wrong.

I did mention two ways to tone it down.

1. Decrease Diffuse_Value everywhere. This would cause all diffuse reflection (direct light-based, direct emitter based, and indirect bounced) to be drastically reduced. The direct and emitter based lighting would then need to be increased to compensate, leaving us with reduced indirect bounced.

2. Decrease IDL Intensity. This would leave direct light-based reflection as is, but reduce direct emitter based diffuse (EnvSphere), and indirect bounced shading (skin to skin).

I actually don't know which is better, but there is a certainty that both of these options are going to reduce the contribution from the emitter lighting (EnvSphere or any other glowing props). Since that isn't the goal, per se, then it stands to reason that one would have to increase the emitter lighting - which is precisely what monkeycloud just explained.

There have been two (subjective?) complaints of late, coming from zanzo, mostly.

1. Environment color (such as sky) is making too much tone change on skin.

2. Crevices glow.

Both of these are addressed by using technique #2 - reduce IDL intensity.

Since I don't know the actual ratios of real life, I cannot dispute nor concur with "zanzo's complaints" (trademark). However, given that you want to reduce those things, reducing IDL intensity will help.

Whether you then want to tweak up the overall lighting (environment value, sun-light value, spot-light value, etc.) is neither called for nor a cause for objection.

I'm very interested in the results you guys are posting. When I have time, I plan to calibrate my camera to neutral settings, and then go outside and measure reflectance on some common objects in shade and in sun, and I'll then be able to tell you what ratio I actually discover. This will tell us something about the contribution we should configure from the EnvSphere, versus the infinite (sun) light. These are, I suspect, the major factors in outdoor lighting accuracy - sun and sky. But that's not the story indoors.

Indoors, I have found that direct light is around 8 times brighter than indirect (bounced) light, and so I am fairly certain that diffuse reflectivity (aka diffuse value) is closer to the neighborhood of 1/8 than 7/8. [I conducted a really shameful test of this. Basically I configure my camera for total manual mode, no change in exposure from one shot to the next. I point it right at a light bulb and also at a piece of paper lit by that light bulb. The ratio in brightness (after remembering to anti-gamma correct the numbers from my photos) was around 8x.]

BB, I have a paint meter somewhere, for duplicating colors, that I think can do that as well. I will have to dig it out and read the book called a manual it came with.

A photo of a ceiling light fixture that is above my desk right now.

It's a compact flourescent light (CFL), surrounded by a white plastic collar.

Notice the ugly color - I hate CFL.

A - direct light source intensity

B - directly lit white plastic (mostly lit by the bulb)

C - indirectly lit white plastic (not lit by the bulb at all - lit by the B area of the collar)

Measure the ratios of these and then take to the power 2.2 to linearize (convert from sRGB gamma corrected values, to straight ordinary values). This tells us the diffuse value, roughly. (Actually the inverse of diffuse value.) Of course, the C area is partly occluded, and the B area is off-center where the bulb throws some less light than is heading to the camera here. But it is ballpark correct.

I get:

A / B = 9.23

B / C = 9.31

Given the occlusion and also the off-center, these ratios are probably closer to 8. Hmmm - fascinating.

Quote - BB, I have a paint meter somewhere, for duplicating colors, that I think can do that as well. I will have to dig it out and read the book called a manual it came with.

Please! Awesome to have some help.

This is using IDL gain set to .1, and the sphere shading disconnected from Alt Diffuse and hooked to Ambient Color with value of 8.
Render gamma 2.2.

I'm liking it, and looking forward to better things to come.

This is getting confusing.

The reason I posted the Environment Sphere (emitter) only scenario was to suggest that the imbalance with diffuse lighting/shading wasn't only a problem with RT lights.

It occured to me since I've been experimenting with this kind of setup for a while, including the EnvSphere plus 1 RT light scenario. In the EnvSphere only setup we see just as much of an imbalance with diffuse light bounce (that's light that bounces off objects as opposed to that which is cast into the scene by the emitter) as when you're using RT lights, or a combination of the two. At least, that's the way it looks to me. The example I posted on page 1 of this thread illustrates this I think.

It's perfectly viable to light a scene only with ambient light in the form of a (good) HDRi, using modulated fresnel reflections as a substitute for specular highlights which we'd usually get from the RT light. The only problem with this approach besides achieving good specular is that of shadows. If we want a definite, sharp shadow then it's still easier to add in a shadow-casting light.

So then we're faced with the problem of having extra light cast into the scene (as well as extra specular if your materials are set up as such). We have an imbalance - and that's on top of the apparent Poser diffuse imbalance when using IDL.

My assertion is that just dropping the IDL intensity to 0.1 doesn't on its own address this imbalance, depending on how you have your scene set up. In the EnvSphere only scenario you clearly end up with way too little ambient light in the scene, unless you increase the ambient intensity of the emitter, and then your reflections get blown out. In the EnvSphere plus RT light scenario you get the right diffuse bounce from that light, but not enough light from the emitter. So what we're looking for in that instance is to find an acceptable balance between the diffuse shader in the material room, the RT light/Emitter intensity, and the indirect diffuse bounce (IDL intensity).

Does that make any sense?

Also, I think the problem of too strong colour bleed/received light from environmental emitters may have been over-exaggerated. In the example of a white object in a green or red room, I'd expect the sphere to be taking on a fair amount of that reflected colour. It might be subject to some of the diffuse imbalance that we're talking about here, but I'd argue not as much as you'd think. If anything I tend to think a lot of Poser renders suffer from a lack of environmentally recieved colour rather than too much.