Liquid Bottle and refraction behavior

[iacdxb]

Hi,

I was converting old perfume bottle to corona... many tries, material and light checking but I was getting strange refraction behavior. I find out in manual... the process glass and liquid method... how it should be.

Strange to me.... liquid should be in-between glass thickness space..., I dont think, in real life it is like this....!!!

If a water bottle with many curve & depths and cuts... it is not hard to keep water inside bottle thickness...?
Is there any other method (no thin refraction) .....?

Thanks.

...

[burnin]

Do show us what you did...

[iacdxb]

sorry that I can not but here is corona method....
https://coronarenderer.freshdesk.com/support/solutions/articles/5000515618-how-to-model-a-liquid-in-glass-in-corona-

Earlier it was glass thickness and then water.... a bit inner, not overlap.

...

[houska]

So are you saying that modelling the liquid perfectly inside the glass is easier than making it overlap a little?

The reason to have it this way is numerical inacuracies (similar to Z-fighting - look it up on Google) - sometimes you might get the correct result, but sometimes, you'll exit the glass and enter air and after a small while exit air and enter the liquid. This is NOT what's happening in real life. In real life, the light exits glass and enters the ilquid. Of course, the result is then different.

[Beanzvision]

I've always done mine this way, as mentioned in the above link.

[Eddoron]

It might not be like that in real life but that doesn't matter in the slightest. It just has to look like in real life.
If we went that route, then one could argue that we couldn't make flat leaves, paper, windows or other thin surfaces.
Infinitely thin surfaces also don't exist in real life.

Different reflection models? Ha! Just model the microsurface and use displacements, not bumps.

[iacdxb]

Thanks for updates.



...

[Cinemike]

I had a deja vu reading this thread and I remember you having asked basically the same question back then already.
Here's a C&P from my last reply there that might help with more complex shapes because it uses the already available geometry and does not involve shifting faces:

"There is another way, though, but it is more complicated than just "normal move" the liquid's surface into the glass:
- Create the glass vessel, create the cap of the liquid, then select the polys where glass and liquid would meet (involves cutting and connecting), and create a selection tag of them (after you joint them with the "cap").
- Assign a glass material to the whole object and a liquid material to the selection."

Hope that helps, if the method is still unclear, I can post an example file.

CU
Michael

[Eddoron]

You're right. He did the same thread. I just thought it was someone else so i checked:
https://forum.corona-renderer.com/index.php?topic=18239.0

[iacdxb]

When meeting deadlines and time is passing in just settings and testings to get nice result and in that I totally forgot that I did almost same posting earlier. Sorry for that.

But thanks for your support/help, fixed and.... happy ever after....!!!

@Cinemike, please drop a example file... I will see the other method.

Thanks.
...

p.s.
in this testings... I found a bug also... I posted in beta page.

[Cinemike]

@Cinemike, please drop a example file... I will see the other method.

Thanks.

Here you are.
I kept the poly count as low as possible to make the placement of the selection as obvious as possible.
Especially with a more complicated inside shape, it should still work accurately and applying an SDS can't lead to faulty intersections because there are none - it is just one object.

HTH
Michael

[iacdxb]

Thanks Michael for that, I will see that.
If you have some time... please see that posting also (beta page), only I am crashing this or it is bug....!

Thanks.

...