model&Rendering: do we need an enclosed space

[smr_VIZ]

Hi, here is an interesting question in my mind for quite a while:

 

First, If I model an enclosed cube, then GI it. I will get one image.

Second, I will model an un-enclosed cube, then GI. I will get another image.

 

In theory, the results will be different since some rays( or irradiance) will get out of the cube and don't contribute to the ovreall irridiance map. But both image look very similar. Any reasons?

 

Here I attached two iamges. First have roof and second one don't. Please pay attention to the top area.

[smr_VIZ]

I am so surprised no one answered this thread? Did I make myself clear or I should explain more?

[Ricardo Eloy]

Well, it will depend on how your scene is lit and the real contribution of that absent face. Also, since VRay's GI is based on pixels, sometimes we get the same result with or without a cap. But, in some scenes, results differ completely. I guess it's a case by case thing.

[smr_VIZ]

Hi, Rick:

 

Thanks for the comments. As you said, I guess if there is a big cap cover the whole scene, then I will see a big difference.

 

Did you have chance to see the model I send you two weeks ago?

 

BTW, are you familiar with MR? I am not if i will get similar results if I use same scene and test it in MR. I might need to try it later.

[Ricardo Eloy]

Hmm... never got the file... maybe you could send it again. Thought you didn't send it...

In theory, you should be able to achieve the same results in MR.