Nested Dielectrics - and how modeling liquid in a glass works

[DMerz III]

Been modeling a lot more in polygons lately, and coming to the conclusion that modeling liquid in a glass and accounting for 3+ surfaces is a tad more difficult with polygons (and subdivision surfaces) than with NURBs/CAD based programs. That aside, I discovered that there are basically 2 methods for how render engines tackle this issue, and it leads to two different modeling methods. This article from Autodesk does a great job of breaking down the differences (see Approach 3 for Keyshot and approach 4 for some other render engines). The latter is using a calculation they call 'nested dielectrics' which is inherently way easier to model, and in my opinion.... animate. (think of a glass filling up...how would you model that using the split surface method? anyone have an example?)

https://www.autodesk.com/products/fusion-360/blog/7903-2/

What are the chances this type of change could be seen in future implementations of Keyshot? Right now, would be a huge wish list request! Is it even possible with the way the algorithm works, or is it a pipe dream for us?


Thanks!

[DriesV]

Hi David,

What are the chances this type of change could be seen in future implementations of Keyshot?

The chances are pretty high. Liquid interfaces are a major area of attention for KeyShot 8.
Any other requests concerning liquids in containers?

Dries

[mattjgerard]

Other than running actual simulations, I can only think of thin film effects on the liquids, which you already have in other materials, and maybe bubble and foam generators, but that is more of a modeling function, so not sure if that is in Keyshot's pervue.

[DMerz III]

The chances are pretty high. Liquid interfaces are a major area of attention for KeyShot 8.
Any other requests concerning liquids in containers?

Dries

That makes me and my team very happy to hear. As for any other requests, there aren't any at the moment that jump out to me. I do appreciate the quick response!

Thanks,

David Merz

[Will Gibbons]

...and in my opinion.... animate. (think of a glass filling up...how would you model that using the split surface method?

Good point. I'm glad I haven't tried to animate liquid in a glass for this reason now that you point this out.