Corona Ray Switcher in glass changes refraction?

[lupaz]

Hi,

I need to have more light bouncing off some glass that you can see here.
So I wanted to use a Corona Ray Switcher for the bounced light only while keeping the same look on the glass.

I'm not sure why the sunlight doesn't seem to go through the glass. Something's strange.


Thanks.

[maru]

What kind of glass material is it? Refractive, thin? Does it have reflection or refraction glossiness lower than 1? Some volumetric properties?
Would you be able to share a simple scene demonstrating this ?

[lupaz]

What kind of glass material is it? Refractive, thin? Does it have reflection or refraction glossiness lower than 1? Some volumetric properties?
Would you be able to share a simple scene demonstrating this ?

See the material attached please.
To send the scene I would need more time. I can't right now. Let me know if you still need it and will send it tomorrow.
Thanks.

[romullus]

How light could go through, if you plugged opaque material into global illumination slot? Your setup doesn't seem to make any sense. If you want more light bouncing off the windows, then maybe it's worth try to turn the caustics on?

[lupaz]

How light could go through, if you plugged opaque material into global illumination slot? Your setup doesn't seem to make any sense. If you want more light bouncing off the windows, then maybe it's worth try to turn the caustics on?

But for refract I left the original material.
I wasn't expecting an opaque material in global illumination to affect refractivity.

[romullus]

The glass is still refractive, it just there's not much what to refract, since behind it is total darkness and that's because global illumination override is opaque. AFAIK, global illumination controls the light rays and refraction/reflection controls visible rays.

[maru]

Yeah, this is the expected result. If you plug a solid material into the GI slot, it will basically block all light (cast solid shadows).

I did not understand this part:

I need to have more light bouncing off some glass that you can see here.

Can you explain what kind of effect you are after? I should be able to help with this.

[lupaz]

Thanks Maru.
What I need to do is have caustics bouncing off the glass. However, turning caustics ON gives me just what it looks like a noisy photon map.
So I thought changing the GI slot to a white material to get more "fake" bounced light.
The effect I'm looking for is like the attached.
The light bouncing off the glass brightens the area in shade.
I could just add area lights, but I was wondering if there was a better way.

[TomG]

Hmm that should work just fine with caustics, I had a test done back when they were released that did exactly this, with no need for any overrides. Can you share what results you are getting with caustics (without overrides), whether your glass is thin, the scale of your scene?

[TomG]

(as a note, best if caustics are not generated by the environment, but by a single bright strong light source such as a Corona Sun only)

[TomG]

As a note, here was my test render where I was experimenting with that very thing (no denoising used here, and not run for very long, just to illustrate they shouldn't be splotchy even without denoising and a decent amount of rendering).

[lupaz]

Hmm that should work just fine with caustics, I had a test done back when they were released that did exactly this, with no need for any overrides. Can you share what results you are getting with caustics (without overrides), whether your glass is thin, the scale of your scene?

Thanks Tom.
My glass is not thin.
The system units are inches.
Corona Sun has "generate caustics" checked.

I'm sending below a rendering showing the bright spots.
Also sending the simplified scene.
I tried 4K and UDH cache.

[lupaz]

The rendering attached.

[Frood]

However, turning caustics ON gives me just what it looks like a noisy photon map.

If you mean just the caustic solver in performance tab, looking at  the screenshot here it seems like you have just not activated caustics in your glass material :)


Good Luck

[lupaz]

However, turning caustics ON gives me just what it looks like a noisy photon map.

If you mean just the caustic solver in performance tab, looking at  the screenshot here it seems like you have just not activated caustics in your glass material :)


Good Luck

I think that's for refractive caustics. I'm looking to get clean reflective caustics.

[Frood]

Ah, ok...


Good Luck

[mferster]

Why not use the caustics render element and then just bump it up in post?

[pokoy]

The rendering attached.

This looks like the sun's direction is too steep for the caustics to be reflected into the shadowed area. You probably know but since you said you want the caustic to bounce into the shaded area I thought it would be worth mentioning.

[lupaz]

The rendering attached.

This looks like the sun's direction is too steep for the caustics to be reflected into the shadowed area. You probably know but since you said you want the caustic to bounce into the shaded area I thought it would be worth mentioning.

Thanks. Yes. Nevertheless, the sample-like splotches a-la vray 2002 is hard to fix in post.

[mferster]

It's not practical in any sense, but you could Isolate just the floor and windows and render those separately you should get less sample splotches

[TomG]

The simple test scene seems to work fine for me. I did move the position of the sun. I turned off the env just to make the caustics clearer in the beauty (otherwise the reflection of the sky drowns them out on the ground in particular. There's the beauty pass and a caustics pass (that I added) just to really pull out the caustics. No denoising used.

Oh I did also delete the render region that was in the VFB on loading, as caustics don't work with render regions.