Sharp shadows of soft highlights?

[Micha_cg]

Hi,

today during a project I run in a curious effect at Vray for Rhino and tested it at the old VfR and at the new version, both show an unreal effect. Maybe it's a limitation of high light calculation, maybe it can be fixed, maybe it works at 3dsmax and it's a VfR bug .. .

The rendering show a reflective object with a white box in front. A point light with "affect specular" only is placed in front, so that there is a soft reflection look at the reflective material. I tested two version of the reflective material, but both show a kind of "shadow" of the highlight of the point light. Can this effect be avoided by a refined Vray core code?

Ciao,
Micha

[svetlozar.draganov]

I am not able to reproduce it here - can you attach this scene ?

[Micha_cg]

Here the Rhino file. Maybe you can recreate the scene at 3dsmax too.

[arcfx]

is there a refraction?

[Micha_cg]

No refraction. It's like the specular effect has a shadow and the shadow can't be soft.

[svetlozar.draganov]

Hello Micha,

The shadows appears sharp because the point light has Radius set to 0 - in other words the light illuminate the scene from a single point and it is not possible to generate soft shadows from lights without any surface.
Here is the same render with radius set to 250:

[Micha_cg]

Hmm, I tested it here and see what you mean. But at my setup with a HDRI env I get a little bit different lighting and so it better show the problem, that the edge of the reflected front object doesn't match the reflection of the specular (the dark "shadow" is from the specular). I think the highlight blurriness is a material parameter and should show the right effect for a small light sources too. It's not a good workflow to adjust the light size for getting the right specular reflection effect. Adjusting the light size would affect the whole scene and can't be adjusted often. For me it looks like there is something wrong in the code. Look at the second screenshot, the reflection of the specular shadow goes upward.

[svetlozar.draganov]

Could you please explain what do you mean by "doesn't match the reflection of the specular" - you are talking about their positions or the glossiness-size?
The reflected dark part is actually the shadow of the light - and the glossiness of that shadow depends on light properties and position.

Here is one more test with the same scene - I have added a new cube objects to better illustrate what I mean.

The reflective material is very glossy and the actual reflections are :

However - reflection and highlight glossiness are independent from the shadow glossiness - and if the size of my light-source is 0 I don't expect to get soft shadows.

Specular Highlights are influenced by the casted shadows onto the objects - and even though they are as blurry as the reflection they are clipped by the shadows.

We got the same thing in nature - if we have an extremely glossy object like asphalt for example the shadows from the sun are very sharp even thought the sun's position and size are huge right ?

[Micha_cg]

Interesting, at your first image you got a wrong reflection like I never have seen it. The mirrored object isn't aligned to the object. So, first we could find the reason for this effect. At my screenshot the red line shows the expected mirrored edge. It looks like the mirror is turned - did you?

And here a part of an image from my project where you can see a sharp "shadow" of the specular effect that doesn't match the mirrored object. You would never see this effect in the real world if the mirror surface has a roughness. OK, I mixed a sharp reflection with a soft highlight, but it should not be so wrong looking. Do you see what I mean?

[svetlozar.draganov]

Interesting, at your first image you got a wrong reflection like I never have seen it. The mirrored object isn't aligned to the object. So, first we could find the reason for this effect. At my screenshot the red line shows the expected mirrored edge. It looks like the mirror is turned - did you?

Yes it looks wrong because of the glossiness - it is perfectly correct according to my opinion.