[Aldaryn]

Its quite difficult to have all faces of a fluid-glass scene coplanar where the two meets.

Just think of an animation of liquid pouring into a container. Without constant boolean operations, and volume checking, there vould be almost no matching faces.
But, if you've got a ray threshold,the matching faces will alvays be in that value. (cant give a smaller thresh than 0...)

A.

[tom]

yeah tolarance is the only solution.. no way!
you can even freeze some strawberry jelly sweet into your glass cup and
watch pulling one side off from the glass with your finger...
carmen can you show a reference photo for this situation?

[Thomas An.]

With all these glass threads I am not sure which one to post this.

I think this is way to go for liquid in glass:

[tom]

circle in the sand ...round and round

[DELETED]

DELETED

[Thomas An.]

circle in the sand ...round and round

Sorry my brain is fried

[Mihai]

Yes Thomas that's exactly the third glass in my tests, but as whiskey said, it's wrong too, the sides have the wrong IOR. Not a big visual difference, but still. The question in this situation remains though for the bottom. Does it have accurate refractions and caustics?

[tom]

Hey I found the solution, let's all come together and open up a business titled "F***ing Dielectrics Inc."

btw/ I found a thick and not-round glass (cubic designed) and i had the glass border exactly as a thickness Can you send a reference photo if you have that kind of glass? the magic is the glass' curvature i think.

[Thomas An.]

Whisky, You are right. As the light transitions from glass to liquid it should be IORglass/IORliquid.

Mihai, I will do some more tests about the bottom and compare them with photos.

-

[Sheik]

There is a potential problem in this approach, witch I explained on the previous page:

So I feel solids would be a secure solution:

Sheik

[rivoli]

just to toss in three more useless glasses, coplanar faces are not a solution, look at this:



from left to right, first galss has three meshes and two IORs (IOR 1,56: glass and inner part of glass this time of red color _ IOR 1,33 water top).
second glass, my beloved three meshes three IORs.
and last but not least two meshes two IORs (coplanar faces where the two solids meet).

[DELETED]

DELETED

[rivoli]

i thought there was an ongoing "mind the gap" water thread in the japanese forum. they are coming to the conclusion that glasses full of water suck anyway.

[deadalvs]

yes, continue with japanese SAKE...

[bader]

rivoli . what is your glass absorbance and adde# and how that effect the seen ?