Forum Groups
  All forums
    Help & Feedback
      Work in progress
      Finished Art
      Non-Max related

Maxunderground news unavailable

Realflow, Gridfluid & object interactions and their distance
show user profile  Ali3D
been working on a scene involving a submarine and a canal of water made up of grid fluid, my problem is when i build the mesh, there is rather a noticeable distance between the mesh and my object(submarine).
anyone knows how to solve it?

read 1457 times
5/7/2013 4:16:55 PM (last edit: 5/7/2013 4:16:55 PM)
show user profile  digital3ds
what are you meshing it with? - if your simulation isn't high enough, then the particles will be too far apart and the meshing will have to be higher = a thick and blobby mesh

The only way to get a really thin and believable mesh is to have a pretty high sim count - if you post some pictures of what the sim looks like (selected) and your sim properties, also what your mesh looks like (and its properties) I could maybe give you some better values to use

it might also help (using grid mesh) to enable "open boundaries" if you haven't already
- Mike Sawicki

read 1430 times
5/7/2013 5:03:18 PM (last edit: 5/7/2013 5:06:46 PM)
show user profile  Ali3D
indeed its so bloby

been playing around with it for awhile, but the calculation is so time consuming :/

edit: i am meshing it with gridmesh

read 1415 times
5/7/2013 5:49:48 PM (last edit: 5/7/2013 6:12:28 PM)
show user profile  digital3ds
ok - so, yea... it's pretty blobby - not horrible

What is this sub doing? just travelling across the surface? use real wave

Your particles look like they're quite dispersed (for this, u might try upping your resolution to something like 5000000 <--- it all depends on what u can get away with, what your time limit is, etc... thats a number i would try next) , but your grid mesh looks pretty good, probably the best you'll get with that type of particle dispersion... Real flow sims are going to take A LONG TIME - that's what your getting into with this program. Sim at a high level for best results, and be super upset when it comes out wrong. The only thing that will make you better is A LOT of fails, u'll get good - I'm talking about like 4 day sims - that's way too much for what you're trying to do; i've had 9 day sims in the past 8 / - also had 2 day sims that were "fails" - gotta suck it up and kick urself

I'm not a huge fan of grid mesh, but in this sit, it works - i'd suggest using the particle mesh (render kit) -If the sub isn't going under - use real wave - use it until the sub "goes under", then re sim with grid domain

if you're submarine isn't doing anything crazy, use real wave...

For production work I usually end up comping MANY realFlow sims together to get the desired result... If you have the renderKit then you'll be in a better position, else everything will have to be rendered seperatley - no prob

So - based on your particle Resolution that mesh is pretty good... try meshing it with the particle mesh (render Kit) - maybe with a radius of .05? hard to say since you're not using it now

[edit] also, your viscosity seems to be way low - normal water @ 60 deg is around .5, are your scales correct? <--- that's a big deal

again - you wont get a much better mesh without upping the particle resolution, I can tell that by looking at how many particles are "over the surface of the sub" - the renderkit mesh might get you closer with this sim - adjust between (.5) to ( 2) relax, and adjust the radius between (.05) to (.0075 <--- my guess) radius, assuming your scale conversion is accurate - I can't remember of the top of my head if the renderKit mesher has a "reach" value, but if it does, i'd put that at like 140 to 160 (that might only be availble with standard particle mesher) this is looking at the sim u have now. lemme know what the end goal is

[double edit] - before you go to bed, try these settings - Resolution, 5000000 (or 10000000 if your cpu can handle it) density 850, viscosity .8, don't be mad if its not perfect, thats what my next settings would be (although I would watch it sim for prolly 2 hours before i moved) - if particles are flowing over your sub, that'll be a good mesh

all in all, good start

[triple edit] "correct scale matters", however, often times in animations you need to "exaggerate" <--- it's one of the 12 principles... how you would exaggerate this would be to make sure your max to rF scale is PERFECT, then, lower the scale of the the realFlow simulation so the waves would almost look like a bowl of milk instead of an open ocean - you don't have to go to that extreme, but hopefully you get the idea. The lower the rF scale, the more exaggerated the waves will be - the higher the scale, the slower the waves will move, and you'll have a longer calc time

[quadruple edit] - not sure how deep your grid domain is, but that will also have an effect... maybe lower the depth (if you can) when you raise the resolution to save sim time - switch ur brain from artist mode to scientist mode and think about the physics
- Mike Sawicki

read 1388 times
5/8/2013 7:08:04 AM (last edit: 5/8/2013 8:43:18 AM)
show user profile  herfst1
Dammit, think I'm gonna have to sit my primary school equvalency test. Tried to follow... gave up.
read 1370 times
5/8/2013 10:42:44 AM (last edit: 5/8/2013 10:42:44 AM)
show user profile  digital3ds
to sum it up - there's no easy button + 5 ways to skin a cat
- Mike Sawicki

read 1366 times
5/8/2013 11:23:27 AM (last edit: 5/8/2013 11:25:36 AM)
show user profile  Ali3D
thank you very much, that was so helpful and in detail, I ran a sim, though since yesterday only some 25 frames :D , but so far it doesnt look that bad, i believe the water slips high so much on the upper surface of the submarine, considering the fact that 10 meters of the submarine is outside the body of water in a still position , maybe i need to rise the friction? though i have no idea how much?

I dont have renderkit, can I still use particle mesh( render kit)? I havent ever worked with it, so I have no idea if i can build it as a mesh the same as grid mesh or not.
and yes, the submarine will go down.

I scaled down the submarine in 3d max 1/100, this way one cube in RF which i have heard is 1*1*1 meter matches the submarine's scale.

and yes, I didnt make a deep ocean to avoid more heavy calculations.

edit: here is the result

read 1345 times
5/9/2013 4:58:36 PM (last edit: 5/9/2013 5:43:17 PM)
show user profile  digital3ds
you'll be able to get a particle mesh(renderKit) into 3ds max from realFlow no problem - you just have to make sure that you are generating the .bin mesh through export central before you "mesh" your simulation, and you have the rF toolbar installed in max - this looks pretty good, especially when i imagine 6 more seconds added onto it, or w/e

your splash calculation seems to be a bit too linear - ie, all of the wave particles turn into splash particles as soon as theytouch the subs wing, this is my opinion

I deff get the feel that this sub is zipping over the surface of water, point made

the particle resolution feels good, nice work, realFlow's a bitch
- Mike Sawicki

read 1322 times
5/10/2013 2:46:34 AM (last edit: 5/10/2013 2:46:34 AM)
show user profile  Ali3D
how should i prevent it from being linear?

I have trouble with splash material, too on vray, have no idea how to make it, reflection and refraction with fall off?

read 1282 times
5/14/2013 3:04:02 PM (last edit: 5/14/2013 3:04:02 PM)
#Maxforums IRC
Open chat window