First off, I'm thrilled that everyone is finally realizing the magnificance of those realistic shadows that a feature like HDRI brings.  I've been testing out the Bryce 6 HDRI and it looks very good, however, let me lay out my thoughts as I see it: this HDRI option is simply too slow for hi-resolution work or animations, and it does not seem to work very well for indoor scenes(unless there is some trick to it).  This brings me to something called Ambient Occlusion. Anyone heard of it?  Ambient Occlusion is an options found in some of the other 3D programs and it creates a realistic shadow pass on all the objects of your choosing, giving you a most fabulous, stunning soft shadowy effect, similar to HDRI, only it's about 20 to 50 times faster than HDRI. So I'm wondering, could we ever get Ambient Occlusion in Bryce, or is that too hard to do.

Yes, I have used AO inside C4D. It is faster but not by much if you have higher settings (which makes a better picture). It can also add some erroneous effects if applied at a scene level. Usually best if used in individual materials but more work required.
One other thing is that AO does not take into account radiosity and can create some lighting errors in that regard. It is a useful tool but one to be used with care.

HDRI option is simply too slow for hi-resolution work

Yes and no, it all depends on the scene. While overall using HDRI slows down a scene because of more complex lighting and especially the soft shadows, I have used it in some large renders (4k resolution), and still been able to render out in under 30 minutes. (but, it is a simpler scene). Complex meshes or lots of pretty glass spheres will make render times shoot way up.

I find just about each scene is different and I have to make decisions on how "cool" I can make the render look without having to wait more than overnight for a finished render.

HDRI might work, but if that is too slow, I may use an faked HDRI workaround, (so I won't have to use soft shadows and then render faster).

Ambient Occlusion would be a great tool to have in Bryce. But, it would just be another tool, and like any other tool, not something to be used every single time, but as an option.

I can see Bryce 7 eventually having ambient occlusion, most likley through a plug-in.

I'm pretty sure that Ambient Occlusion would be very useful in just about every scene I can think of(except flat open outdoors).  Example:  A bed in any scene  makes everything under the bed dark, regardless of your other light settings.  Or the rough outdoors, where the bottom of a canyon would be darker than at the top; or under a stone arch or overhang where it would be darker than elsewhere.  These natural shadow effects make all the difference in the world and can turn a flat, fake-looking scene into  realism that's hard to imagine until one sees it.
Considering how fast Ambient Occlusion can be, I think noone should live without it.

The theory behind AO shaders is pretty simple, the mathematical implementation shouldnt be a problem for a programmer who can program HDRI solutions. So in that respect, it should be possible to put it in bryce.

However, it's defenitely NOT 20-50 times faster allround. It might be very much faster on a simple object, but the more detail u have in your scene or the more data, the more you will see rendertimes increase and you might want to increase the quality too for complex geometry, which boosts rendertimes even more.

AO can be usefull yes, i do use it myself, it's easy to give extra definition to shapes, it's good for faking some natural lighting effects, but it's not always the best solution per se.
AO is a render solution that calculates the amount of occlusion of a point of an object by other surfaces of the object itself and surrounding objects relative to a hemisphere around the scene. This creates a totally uniform occlusion effect that doesn't account for different propperties of different surfaces in terms of light interaction and it ignores actual scene lighting in it's entirety, including color bleeding, caustics, refractions, 'diffuse light bouncing'. By lack of this, naturally, you dont aquire a lighting solution that will always fit all your needs.
In the dark bedroom example for instance you might want a bright streak of light comming from behind curtains, or a slightly opened door, and you will notice AO won't make the light from the bright streak bounce off surfaces and you'll lose the nice soft glow of it that would in real life be cast onto surrounding objects. Shadowing onto surfaces around the a white ceiling would in real life be less intense then on surfaces surrounded by dark objects, AO doesnt do this for you either.

I will also say the examples I have seen from researching AO are some very nice eye candy. I look forward to (hopefully) having this in Bryce in the near future. It is a another step towards a little more realism.

Yup, it's a handy tool, no doubt about it. I just wouldnt put my money on it being the hallelujah tool for total realism and fast render times :)

I guess the bottom line really is this:  how do you achieve those neat real-light effects in a useful and practical way.  A 90 hour render is obviously not a pleasant option no matter what the final result is.  One of my full size renders using that "volumetric world" option, for example, would take about a thousand hours to complete, so I never use it.

Now I have to go and figure out how to successfully use HDRI on an indoor scene, with ceilings and all.

Yeah, basic rule #1 - never use "volumetric world".  ;o)

HDRI - pretty much moot if you have an indoor scene with walls and ceilings, imo. That kind of scene wouldn't benefit from hdri. I mean you already have a full enviroment for reflections, trnsparencies and refractions. And, in real life, in an indoor room, there would be lamps/candles and such somewhere, just place spots and radials there.

Quote - Yeah, basic rule #1 - never use "volumetric world".  ;o)

HDRI - pretty much moot if you have an indoor scene with walls and ceilings, imo. That kind of scene wouldn't benefit from hdri. I mean you already have a full enviroment for reflections, trnsparencies and refractions. And, in real life, in an indoor room, there would be lamps/candles and such somewhere, just place spots and radials there.

Not having Bryce in my toolbox of applications, I can't say for sure, but you may be able to utilize HDRI to simulate "bounced" light in an enclosed room.  Or perhaps Bryce already has light bouncing or another way to fake GI for such things.  Note that simple GI is different that HDRI in that GI can be used as a way to bounce photons around a 3D room from existing lights in a scene..

For example, you could theoretically make your own simple, custom HDRI to "compliment" the indoor lighting setup you have in an indoor scene, and use it to simulate secondary light bounces from your light sources, and give the scene some soft ambient (AO-style) shadows..  That is, if Bryce doesn't already have another way of doing this that's more accurate/direct.

A better example of this would be to place one radial light in a room.  Normally, the light will end where the falloff ends, and the rest of the room will be black.  Plus, you won't get soft ambient shadows where you would expect them, because there's no light bouncing around in there from your single light source.  A low-intensity, solid-grey lightprobe could be used as "fill" lighting to provide the faked light bounces, instead of adding a multitude of standard lights.  Of course, it will only work if you can use HDRI in an enclosed scene in the first place.  If Bryce impliments HDRI like a skylight, then you'd have to take the roof off your room in order for the probelight to get in there.  This would actually be the kind of situation where IBL/AO or standard GI (photon mapping) would come in very handy.

In the mean time, a description of my test setup.  Booleaned cubes to get an empty box - a small sphere in the middle, camera inside the box.  The walls are MadMax's glossy white HDRI mats, floor is glossy black HDRI mat, sphere metallic car paint (red).  No ambiance in any of the materials.

Atmosphere off, sun disabled, no extra lights.

Render settings are Premium, 4 RPP, blurry reflections and transmissions (to get any stray HDRI effects showing up).

Wireframe.

The render with HDRI is much more interesting (and quite slow at these settings)...

Here's the HDRI render done quickly for first impressions (quality 0, regular AA), I'll post the premium render when it's done.

Quote - Yeah, basic rule #1 - never use "volumetric world".  ;o)

Nah, it's not that bad. For instance, my After the Rain was rendered with volumetric world. Yes, it did render for a long time, but it was not very much different from other renders. > Quote - HDRI - pretty much moot if you have an indoor scene with walls and ceilings, imo. That kind of scene wouldn't benefit from hdri. I mean you already have a full enviroment for reflections, trnsparencies and refractions. And, in real life, in an indoor room, there would be lamps/candles and such somewhere, just place spots and radials there.

I beg to disagree. For instance, you've got a lamp in the corner. In the real world, you get the lighting both from the lamp and the light bouncing from the walls. That kind of lighting is very difficult to achieve with just Bryce lights. As to more realistic lighting, I think I'll try something with the radiosity simulation on http://bsmooth.de/BSolutions/#Radiosity. Yes, it means another long render and it means layering and blending in Photoshop, but we'll see.

The question becomes - Is this an artifact of the particular HDRI Image I was using, or is it a common effect?

You've either linked or uploaded only the black image...

I see the sphere (dimly), wall, and reflection in the other wall.  Two different computers, firefox and IE.

Or am I going bonkers?

Too dark for my monitor. I only see something dimly when I raise the middle input level in Photoshop from 1 to 1.5. So either my monitor is too dark or yours is too bright. Yes, mine is dark, but not that much.

Monitors aside, it's still interesting that HDRI works (however dimly) in a totally enclosed space.  (I'll also be checking my monitor settings)...

With my freshly calibrated monitor, I can just make out basic division between the walls and a hint of sphere.  So HDRI inside appears not very useful, but could work (somehow).

Well, yeah, now I see what you mean, but only when I enlarge the pic to something like 160%. OTOH, that's not what I (and I think AS) meant. I am talking about HDRIs representing an enclosed space, not about HDRI being visible in an enclosed space. In fact, I think what you discovered is a bug. If the HDR is completely outside the box - while your sphere and the camera are inside - there shouldn't be anything visible in the box. For my example, see St. Peter's probe and Grace Cathedral probe from debevec.org or check http://www.worldof3d.com/tutorials/tutorial8-01.html for the panorama.

ok, there seems a bit of discussion about HDRI based IBL and indirect lighting solutions... There's a litte difference between the two which i feel i should point out.

IBL is image based lighting; Light cast in accordance to a the brightness and color of an image. For scenewide IBL such as in bryce6, the light will be cast from what is effectively a spherical dome around the entire scene.
Added benefit of HDRI for IBL is that there is a more then 256 degrees of brightness in the image used as light map, thus enabling you to create more realistic contrasts between for instance a sunlight and a lightbulb (both might look white to the eye but their brightness is difference).
IBL is not indirect ligthing, it doesnt calculate light bouncing off one surface onto another. Thus the light an HDRI might cast through the windows will not bounce around the room. So, as a result, the scene will not really benefit from it too much.

So as a result of this, which AS already stated, an interior render already has an environment for reflections etc. An IBL scene would mostly leave the inside of a room unaffected, after all, the walls and ceiling will obscure any light cast from the dome around your scene.
The only light from the IBL getting into the room would come through the windows. Once there, it will only be able to cast into a small part of the room because it won't create any indirect illumination yet. You'd need TA for that part of the solution. (or you could simulate it with textures and strategically placed light-sources)

An outdoore scene benefits from HDRI based IBL because there's light from the 'skydome' comming from all directions siulating light bounced off buildings and other objects surrounding your scene as well as sunlight diffused throughout the atmosphere and, being an outdoor scene, there's nothing to entirely (or at least mostly) obscure your image's subject from this light.

Ray - All true, so Erlik is right and this is a bug?

Well, I'm not sure what's causing the light inside your setup but i doubt it's indirect light. Even if it is indirect light, it should not have been there in an enclosed and completely unlit environment.

There's a theory i can come up with though..
We know from TA experiments that there's a problem with the brycean cube's surface normals that make one of the cube's faces react strangely to light, maybe this affects the HDRI example scene too?

But yea, I think it should be a bug. Light, direct or indirect, should not be inside an enclosed non-transparent environment with no light source inside it.

IBL (a la Bryce) is good to doing  figures/items (CG) to be composited into a room (real) image from which the HDRI (photographic) was created. This was the movie biz's original and primary usage of such lighting.