Light through Glass Material?

[MegaPixel]

Can it work that way, or do you have to seperate your glass objects and adjust their properties to make them transparent or invisible to the Light source? Affect Shadow/Alpha doesn't appear to be doing the trick for me allthough I can see "some" glass shadowing, I'm not really getting the correct amount of light coming through the glass.

I'm using a Vraymtl for my glass with the following parameters:

Diffuse
-Diffuse: Black

Reflection
-Reflect: 0/0/75 HSV (about 25%)

Refraction
-Refract: White (100%)
-IOR: 1.1
-Max Depth: 6
-Exit Color: White
-Fog Color: White
-Fog Multiplier: 1.0
-Affect shadows: On
-Affect Alpha: On

BRDF: Phong

Options
-Trace Reflections: On
-Trace Refractions: On
-Double Sided: On
-Use IR Map: On

Thanks in advance for your help - Dave

[Clifton Santiago]

With Affect shadows checked you should be able to see plenty of light through your glass, as it is essentially a fast, fake caustics. Are you sure you don't have anything blocking the sunlight? Like a planar Vray light in front of a window? I have discovered that even if invisible, the vray planar light will block other vray lights.

[MegaPixel]

http://www.cgvisions.com/temp/vray_glass.jpg

Thanks for the reply Clif. The image link above should give you a better idea of what I'm looking at. There are some areas of this glass object that are behaving correctly and some that are not. Where the glass sheets are ontop of the shelf cabinet material, I'm getting a strange flat black return. Towards the back of the shelf, I'm thinking I should be seeing a brighter wall for a generic clear glass material. There are no Vray lights in my window openings to obstruct the ray paths either. Any other thoughts? Thanks

Dave

[RErender]

looks like the glass is coplanar with the cabinet,i'd try moving it out a bit. If that doesn't change anything you might try increasing the depth in the glass mat.

[zeronove]

It seems to me an issue related to faces sharing the same space; try to move the glass a bit away from the cabinet or change the value in the secondary ray bias

a.

[Clifton Santiago]

Okay, I see. As for the "WTF" part, and even the "Looks Dark" part, maybe you have co-planar or coincident faces, like two pieces of glass exactly on top of one another. Both clear and opaque materials will render black in that case. With clear materials you should have a space or an overlap with any other geometry.

Also, do you have GI caustics: Refractive checked? You might need that for ambient GI to pass through the glass (though if you have affect shadows checked I think it is overridden anyway).

[Clifton Santiago]

Oh, sorry, my post says the same as above... I took to long to write it

[MegaPixel]

http://www.cgvisions.com/temp/vray_glass2.jpg

Ok, I've posted a new shot. I went ahead and modeled a basic glass tower shelf from scratch that has no coplaner faces - all parts are atleast .01 inches apart. The interesting thing is, the two cabinets on the left share the same exact material. Their proximity to the wall and other objects affects how dark they get. The tower on the right just has a basic vraymtl on it to show you what the geometry looks like. The glass material settings are also in the above image. I've tried an aweful lot of options within the material and rerendered quite a few times only to see the same phenomenon happening in the corner.

Rerender, I upped my max depth to 8 (not much) but I guess I could try upping it to like 50 to see what happens.

Any other ideas? Thanks - Dave

[dapeter]

Have you tried rendering without reflection (reflection turned to black), or rendering with the reflection turned up to white with Fresnel enabled?

A quick test I did showed that having any reflection on the surface blocks light coming through it (as it would in the real world). If you need some reflections to show, try enabling Fresnel, you'll get a more realistic glass anyway.

[MegaPixel]

Well, I think I may have found a way to fix my problem and that requires excluding the glass objects from the GI calc. The images below show the results with refract only and then fresnel reflections:

http://www.cgvisions.com/temp/refract_only.jpg

http://www.cgvisions.com/temp/fresnel_reflect.jpg

These images show the object completly excluded from the GI. Maybe I'll try just lowering the receive/generate numbers a tad instead of completly excluding it since the glass does need to interact with it's environment. Thoughts on this?

Dave

[studioDIM]

maybe you have refractive caustics turned off in the IRmap settings?
blind guess, hey...


Lele

[jacksc02]

megapixel, check the max depth of the glass material and also the max depth in "global switches"

[DaForce]

Yeah i would say make sure refractive GI caustics is on (in the GI rollout) and then change your refract exit color to a bright red and make sure the little box next to it is ticked.. then just keep uping the bounces until most of its not red, then untick that box.

[MegaPixel]

maybe you have refractive caustics turned off in the IRmap settings?
blind guess, hey...


Lele

I did have refractive caustics turned off. I thought caustics were a "bad" thing. Here I went through all this trouble to make sure everything in the scene wasn't casting anything remotly related to caustics to help my render times and now I come to find out I need them on for refractive glass transmission? Ok I'll turn that back on and see what happens.

Daforce, I allready have 8 secondary qmc bounces goin for me. As far as max depth goes, I turned it up to 24 and still saw no change.

I was hoping this would be the simplest most generic glass possible to reduce render times but it would appear that in order to get light to behave correctly, caustics need to be involved. Bummer

[DaForce]

These arnt caustics like you think they are. They wont affect your render time much at all I reckon.

See how you go and post results