Liquid seems bad

[Zvi]

The liquid in the bottle seems bad. I'm attaching screenshots with the liquid and the glass settings.

[Chad Holton]

What does it look like with the glass hidden and a polished metal applied instead of liquid?

[Zvi]

What does it look like with the glass hidden and a polished metal applied instead of liquid?

Like this.

[Chad Holton]

Not quite sure, but it looks like it needs to be smoothed out in the modeling software or something. What modeling software was used to create this?

[Speedster]

Is there a tiny "gap" between the liquid geometry and the inside of the glass?  That's mandatory.  Can be very small, like .001" or so.  If both surfaces are in intimate contact, just scale down the liquid body by about 0.995% or so.
Bill G

[richardfunnell]

Your best result will come from modeling your scene using surfaces, not solid bodies. Or if you do model with solids, break out your individual faces to get the most control over your surface IOR values for the front and back.

Adding a gap between your glass & liquid will give you an incorrect result, since that doesn't exist in real life. I'm attaching an image I found on the forum some time ago (apologies to the original poster for missing credit, I can't find the thread!) that helps break the process down.

This was also a topic specifically addressed in a webinar for transparent materials, you can watch it in the link here:
https://youtu.be/HlxF9peYod0?t=33m52s

[Zvi]

Not quite sure, but it looks like it needs to be smoothed out in the modeling software or something. What modeling software was used to create this?

I'm using Maya. Why is it related to smoothing?

[Zvi]

Is there a tiny "gap" between the liquid geometry and the inside of the glass?  That's mandatory.  Can be very small, like .001" or so.  If both surfaces are in intimate contact, just scale down the liquid body by about 0.995% or so.
Bill G

Yes, there is gap.

[Zvi]

Your best result will come from modeling your scene using surfaces, not solid bodies. Or if you do model with solids, break out your individual faces to get the most control over your surface IOR values for the front and back.

Adding a gap between your glass & liquid will give you an incorrect result, since that doesn't exist in real life. I'm attaching an image I found on the forum some time ago (apologies to the original poster for missing credit, I can't find the thread!) that helps break the process down.


Thank you for the important information. This scene is based on the tutorial you recommended. That's why in the outside refraction index of the liquid I put the value 1.5. There is no reason to break the model of the liquid to parts if it is wrapped with the same material - glass.
This was also a topic specifically addressed in a webinar for transparent materials, you can watch it in the link here:
https://youtu.be/HlxF9peYod0?t=33m52s

[Chad Holton]

At first, it looked unsmooth but after looking at it again, I think you have some normals flipped.

[Zvi]

At first, it looked unsmooth but after looking at it again, I think you have some normals flipped.

You were right. Thank you.
And thanks to everyone who bothered himself offering his help.
Even those who claimed that the problem was the gapless were right. There were places of contact between the liquid and the glass.
Again, thanks to everyone!

Now it looks like this.