I'm starting to mess around a bit with bagginbills environment sphere. I read through the instructions and understand the basics. But there's some things that are confusing me. I'll number them for better readability:

1. I easily get this render on the bottom:

https://sites.google.com/site/bagginsbill/free-stuff/environment-sphere/environment-sphere-instructions

But i notice there's no shadow on the ground. In his other renders on this page:

https://sites.google.com/site/bagginsbill/free-stuff/environment-sphere/environment-sphere-demo-renders

there's shadows on the ground. How to i use the sphere to get shadows on the ground? I added one infinite light for the sun, and i could get shadows on the body, but not the ground.

1. When i turned the shadow strength from 1 to 2 with GC enabled, the shadows came out blue. I just wanted to darken them. Any ideas what's going on here, and is there a way to darken the shadows without that weird effect?

2. Say i want to use a particular scene or some scene props, with a ground plane and maybe a few walls or rocks. How do i set up the sphere so it uses the props, and includes a sky as the picture for a prop based outdoor scene with a sky? Would i use one of the EvoDome or EvoHemisphere that's included in the prop folder for this?

3. For the HDR or LDR image to come out right, it says to decrease the camera focal length from 75mm to about 15-30mm, otherwise it will be too blurry. Is there a way around this, or do i always need to decrease the camera to those settings?

4. Where can i find some good sky pictures for the skydome?

5. How would i apply indirect light to an indoor scene? Would the IDL bounce rays off the insides of the scene, or would i still need to use the dome?

I'm liking indirect lighting a lot. With one yellowish infinite light, it makes an outdoor scene look really realistic, much better than using regular lights. Any other tips would be helpful. I understand basic lighting to a decent level, but i'm still new to using bagginbill's sphere and how it works.

poser ground plane has shadowcatch-only setting, but it's not very good IMVHO, and may not be useful with IDL-only renders.  bill's free shadowcatcher shader is much better for poser ground plane, but it will be difficult to learn in most cases, and it gives unexpected results when using hi-res jpg background and HDRI lite source on the sphere (e.g. sIBL method), due to the refractive nature of bill's shadowcatcher shader.

Answer to queston 3: Depending on the scene's ground plane.  If you use a plane mesh which has mountains, dunes, and mounds then the dome would be most satisfactory for this particular scene strictly because plane height would hide the ground and dome seam.  If you are using flatter terrain then the hemisphere would be more appropriate.  Using the entire sphere would involve utilizing an image which may contain a 360 degree ground and sky or predominantly sky or interior.

Answer to question 4: To still use a narrower focal length (i.e., 75mm) you would require a very high image resolution which I have forgotten atm.  However, these high resolutions can bog down your system and make Poser slow to respond.  Decreasing the focal length will widen your field of view or incorporate 'more' background (viewing more pixels) into the scene sharping the image. 

Answer to question 6: If your interior scene incorporates windows, a dome will add more realism to the render.  However, ambient value must be increased to a rather large value for IDL to illuminate interiors correctly and realistically.  500% or higher sounds about right.  Now, if your generation lighting from fixtures, candles or torches, inverse square point lights would be adequate.

I wonder if it would be better to just use a ground/scene props then to capture the shadows, and the dome on top with a sky planted on it, rather than what was mentioned on bagginbills site. I curious though how he got the shadows to appear on the last few renders here:

https://sites.google.com/site/bagginsbill/free-stuff/environment-sphere/environment-sphere-demo-renders

Looks like it was possibly done by rendering shadows separately then using photoshop.

... Looks like the ground plane set as shadow catch only to me. It seems unlikely that BB would use compositing in one of his example renders.

No Photoshop.  Everything BB creates is within Poser.  He used his own shadowcatcher program to introduce those shadows.  IIRC, there is a thread about it's use in the forum.

Is there a place where i can find some decent free HDR or even LDR files for backgrounds and skies? And for LDR files, what image resolution and size should i be looking for?

And one other thing i noticed. Here is a test render with the focal at 30mm. Michael 4 looks good in here, but the background is a bit blurred. So i decreased the focal to 15mm, as shown on the second post. The background looks good here, but michael looks a bit weird at this focal. Is there a way to fix this?

Anything below 35mm focal length goes into the realm of wide angle photography.  You will begin to see fisheye distortion with values 24mm and below.  There is really no fix for this type of effect except angling your camera view.  Avoid doing a tilt view from ground up to sky, architecturally, structures will look like they are falling over.

Ahh, ok. So i probably should be looking for images around 8000 x 2010 then, and for smaller ones, apply sharpening and dof techniques in photoshop, so i can keep the camera at 75mm. 35mm doesn't seem bad either, but anything below that, everything looks distorted as you mentioned.

Here's another quick test render where i applied a dof effect in photoshop. Took about 5 mins, but shows the potential of this indirect lighting with bagginbills dome. Not to mention, it also saves me ram since i don't have to add any props or scenes. Just replace the default m4 in your mind with any character, and it would be a decent render. I like how you only need one light and the dome does the rest, saves a lot of time with more complicated light setups where you have to render and re-render to get the right lighting and shadowing.

Is there any other tips or recommendations you got for using bagginbills dome?

The render I posted only has one infinity light reduced to an intensity of 30% and slightly tinted yellowish red.  When using IDL and a dome, keep the lighting simple especially for outdoor scenes.  IDL will aid adjacent surfaces to reflect additional lighting into the surrounding area.  Also observe how natural lighting behaves in real life.  Using this knowledge will help you determine the best placement of lights and what to expect.  The same applies to indoor lighting. 

Quote - ... Looks like the ground plane set as shadow catch only to me. It seems unlikely that BB would use compositing in one of his example renders.

BB has a shadowcatcher shader... 

https://sites.google.com/site/bagginsbill/free-stuff/shadow-catcher

Yeah, i just read it, but it looks pretty complicated and time-consuming to use. I think i'll take the photoshop route on this one, or use other techniques such as rendering shadows only, then manipulating it within PS, or placing a prop on the ground and make it fit well within the photo, and grab the shadows on there.

hborre: I have a few scenes where it's a room with 3 sides, but missing the ceiling and front so light can get in. What i did was use bagginbills dome with the doge2.hdr file to light up the scene. I kept the camera at 75mm because the room walls from my angle blocks the doge2 image completely, so it doesnt matter if it's blurry or not. But for this case, i used two lights for the scene, for shadow purposes. Everything lighted really well. For indoor scenes, is it ok to use multiple lights with the dome/IDL to get the look and shadows we want?

This could also be done by applying that same doge2.hdr file to a diffuse IBL, but i noticed the shadows didn't come out as good as with IDL and bagginbills dome.

That is a difficult call.  There are many older half-room models around which make set lighting a nightmare where IBL and IDL are concerned.  Some users complete the rooms by adding flat planes to take advantage of the IDL feature.  Others use a darkened dome to simulate the same thing.  And becomes even tougher when a daytime image is used on the dome and your room is incomplete.  If I had to chose between IBL and IDL, IDL would be my preference.  I have found that IBL ambient lighting is too far reaching; it must be used at very low intensities or it will overpower your shadowed areas.  It is very indiscriminant. 

For indoor scenes, multiple lights is perfectly acceptable.  You are trying to simulate an interior stage which may have several light sources for illumination.  The purpose is to duplicate real lighting situations.  Try decreasing the ambient value of your dome so it's intensity won't interfere with your stage lighting.

picco, a lo-res diffuse IBL might be enough to get nice soft lighting with those highly detailed 2-wall-and-floor sets.  the IBL can have shadows but several years ago, when they were using IBL instead of bill's envsphere, they would use AO to fake the occlusion shadows that are calculated with IDL.  you could ask somebody who is still using poser 6 about IBL/AO.  otherwise you could try the sIBL method, with IDL enabled, to get ray-traced occlusion shadows. for one of these poser renders where there's no reflection details, even if indirect specular has to be simulated, a lo-res image as indirect light source may suffice.

I daresay this can be confusing: IBL, HDRI, IDL, AO, shadow-map, ray-trace.  it takes time to learn.

hborre: Yeah, i tested out the half room with both IDL and IBL. Even though with the same HDR file attached to them brings similar results, the IBL does wash out shadows whereas the IDL gives some very nice, realistic ones with the 2 lights i added for the room. This is easily becoming one of my preferred setups, because not only does it give me full control and brightens the scene well, but i can leave gamma correction ticked off. GC is something i've learning to hate very quickly =P Thanks for all your help man. I learned a lot from this thread. I'm curious, do you usually render with Gamma correction on? I used to always leave it ticked on before, but now i prefer to leave it ticked off since it tends to make images look flat and dull, killing highlights and shadows.

Miss Nancy: Yeah, lighting has a very high learning curve, and there isn't one comprehensive tutorial out there that explains them all in detail. I had to use multiple tutorials and the advice of other people to finally understand lighting. I pretty much understand IBL, HDRI, AO, shadow-map vs raytrace pretty well now, the only thing i had to learn was how to use IDL. And with the help of you guys, i have a good understanding of it now =)

I always render with linear workflow, or Gc on in the render settings.  It simplifies the lighting as found in real natural surroundings, which means that where older Poser versions required multiple lights to brighten a scene, I can achieve the same thing with fewer lighting resources.  The render I posted, 1 infinite, low intensity and sky dome rendered with Gc=2.2.  The shaders for Miki 4, the hair and the plant were modified using Snarly's EZSkin script.  Everything was changed to suit linear workflow conditions.  This means reducing diffuse_values to 85 percent, changing diffuse_color chips to white, converting the gammas of all data maps to 1 and adding SSS to all organic models in the scene.

Ahh ok. So in order to use gamma correction, i need to make those changes so it corresponds to linear workflow. I kinda like the look with GC, but i can produce the same results using IDL or a diffuse IBL to brighten the scene, so i'm not sure which is better suited for me. Do i have to make any changes to make EZSkin 2 look good with Gamma correction ticked off? Anything with Ezskin 2 looks good with GC enabled, but i haven't really tried it without GC yet.

Currently, EZSkin should work equally well in either situation.  But clearly understand, lighting plays a very intricate role in how the scene appears after rendering.  Gamma correction will brighten the scene, making it possible to discern shadow detail that was not possible in older versions of Poser.  Our eyes are more accustomed to see better under darken conditions than bright.  In non-linear workflow, users either added more lights to the scene to simulate what we naturally perceive, or add cheats to the material room to enhance textures to that stage.  Diffuse IBL adds ambient lighting based on a projected image, but once you add stage lighting (infinite, point or spot) you amplify the light intensity which looks correctly in a non-linear environment but becomes completely washed out when correction is applied without adjustments.

A linear workflow works better for me; I don't need to over think my lighting nor do I need to add multiple lights needlessly.  I alway strive for something that is very natural and realistically possible, not create an intentional imbalance unless I looking for dramatic overkill.

Sweet. It all makes sense to me now. No wonder why my earlier images looked washed out. I tend to light with multiple lights because i go for certain effects, whether it be multiple shadows, colors, etc, so i guess for me it's better to go with non-GC. Before, non-GC was too dark, but ever since i discovered and learned IDL with bagginbills dome, and using HDR in diffuse IBL's, it's no longer a problem for me. The one thing i do like about GC though is it makes postwork easier. It's easier to darken shadows than trying to brighten them up, since nothing is there to brighten up. I guess it all comes down to preference. But it's nice to know that ezskin looks great under non-GC, since now i got both options with it.