[Aniki]

well, what is it exactly you want to do ? animation or still ?

in both cases you'd need the maxwell materials to use the uvw data to properly texture those alpha maps. In the case you use RealFlow, you'd also need the plugin to work with the animated textures applied to the mesh geometry. also complicated and "error prone"

another way would be to try having the mesh from inside realflow already built correctly considering clipping objects.

[simmsimaging]

I only work in stills so that is my only concern. Easier in some respects, but far less forgiving in others

Anyway, it's not possible to get a perfectly adjusted mesh out of Reaflow, because not matter how perfect the fit to the container, you still have the problem of the interface, so some part of the mesh needs to be removed from the meshes to create the proper refractions.

Same with the drops. Maybe it's not such a big deal if the refractions don't work "perfectly" for droplets, but for liquids in a glass it can definitely show up more obviously.

This is not just a Maxwell issue btw: the only render engine that gets around it, AFAIK, is Vray. I have the same problem with mental ray, Octane, Fry etc. etc.

/b

[Aniki]

vray... we call it fakeray here

did you try with the clipping objects in the mesh tab in RealFlow ?

apart from that I really doubt there will be an optimized "one click" way of doing this. always depends on the scene, what is actually happening etc..

well, whishing you luck and success whatever route you take and looking forward to the results !

hope I could help you on the issue though.

cheers

Aniki

[itsallgoode9]

I think i know a way to do this which is fairly simple. I don't want to jump the gun, so I'll post some samples once I have time to test.

[simmsimaging]

I think i know a way to do this which is fairly simple. I don't want to jump the gun, so I'll post some samples once I have time to test.

Please do - I just jumped back on the Maxwell bandwagon with the current sale on licences, so I'll be very interested in how it works out!


thanks /b

[itsallgoode9]

Sooo, this is what I was trying, but I cannot get it to work for some reason. I feel like i've tested this in the past and it has worked, but now the inner liquid just disappears when doing this. Although I know I"ve seen some members test this and it worked out. I know it's not a correct way to do the glass, but it seems like this is the simplest solution for this liquid issue (if it will work)



here is the result that one member was able to achieve with this setup:

[simmsimaging]

That is the method for Vray, but I don't think it works in other engines (not the one's I've used anyway). If someone got that working in Maxwell then that would definitely solve both types of problems (drops and interior liquids). Maybe Tom or Mihai could chime in on it?

/b

[itsallgoode9]

So, I figured out what was wrong, I had to REALLY adjust the attenuation depth to get the color to show up. It's not perfect by any means and you can see there are some weird coloring and banding artifacts, but it might work well enough as a hack.

Modeled the correct way:


Modeled the VRay way:

[Aniki]

nice try, yet the black bottom bothers me. you might also see this is wrong when animated

[simmsimaging]

.... Maybe I think the black area is problematic, but that could be adjusted in post. What's more of an issue is the refracted light and caustics that are either gone entirely or weakened. It's the effect on lighting that will kill it IMO.

NL: is there any remote chance of adapting the way Maxwell deals with interfaces? Maybe a special purpose material like AGS, but for intersecting dielectrics etc?

/b

[simmsimaging]

Been working a bit on the boolean approach, but realized I'm not sure how to set up for a pour

I have the Thomas An diagram for liquids seen inside of a glass:



But how do you set up the surfaces properly if viewed from *above* the rim, and with a pour hitting the liquid? I can merge the pour and liquid mesh together with a boolean operation, but I'm not really clear on how I should be setting up the glass mesh once I'm looking down on the top surface? This is a rough idea of the setup:



Tom - can you clarify?

Thanks /b

[itsallgoode9]

Which software are you working in for modeling? Have been milling over this all night and have an idea of how the boolean approach can be done in Maya...assuming the booleans in Maya actually work, which probably has a very low chance of happening with geometry this complicated.

[simmsimaging]

Max 2011/2012 for me.

[Aniki]

Send me the objs and I will set them up for you in the correct way.

That diagram is wrong, as the glass doesnt need to "contain" the fluid. I dont even know if that infinitesimal rule applies to maxwell.

The pouring from the view above is kind of the same setup I used to do. And it worked correctly that way. Even with colorchanges in the glassbody according to the liquidcolor.

No darkening etc.

On the left just the glass without the substracted liquid mesh.



Important for the liquid body to have the glass geometry normal reversed before boole/merging so the resulting liquidbody has all normals pointing outwards.

[tom]

Aniki, what you do is OK, except I don't think you should worry about the direction of normals. It should render the same look even with the normals facing opposite direction.