Metas for water effects

[NickHutson]

HERE is my illustration (FINALLY) of the main idea of how Meta objects could work for the creation of water effects. The rest of the Idea can be found in this stream:

 

http://www.hash.com/forums/index.php?showtopic=14212&st=60

 

 

 

Complicated or not the work would be well worth it. Just like the work put into making cloth was worth it. The out come would send our projects over the roof!!

[aaver]

Sorry, but I still don't understand the advantage of your idea over A:M blobbies.

[Roughy]

bloobies take forever to render...

[jamagica]

with metas you can create a model and basically, by hand, control each bubble you make. The meta balls are actual models, rather than a material of bubble simulation

[aaver]

bloobies take forever to render...

I see no fundamental reason for meta ball geometry to be faster than blobbies. As a matter of fact, I think they are basically the same.

[John Bigboote]

Great idea! As I see it, the 'meta' balls would not only be hand-controlled (who would want to hand animate 1000's of units?) but mainly DYNAMICLY controlled as in rigid bodies...thus allowing them to 'flow' amongst themselves. They would have adjustable properties for...gravity....repel/viscosity...bounce... etc

 

This WOULD be a major addition along the lines of hair, cloth... so unless a highly talented programmer steps forward, I imagine we won't see anything like this until V13,14,15....

[aaver]

with metas you can create a model and basically, by hand, control each bubble you make.

When it comes to realistic water built of tens of thousands of particles, that is hardly an advantage, is it?

 

The meta balls are actual models, rather than a material of bubble simulation

But is that an advantage? As I understand it, blobbies and meta balls are only two different ways of implementing the same thing.

[NickHutson]

The way you understand it has no fundamental value. If others shared your reasoning we would never have even seen Perfect Storm.

 

John Big Boote however gets the idea. Blobbies are very good for making soda pop. The only problem is that they DON'T morph together like real life water particles do. The illustration above displays this, which is the very basic concept of how the morphing of one side of the object would work. Get thousands of these doing the same thing that are also controlled by gravity, object collision and velocity and you get pools and water falls.

 

Forget rendertimes and compexity. If we were worried about things like that we never would even have hair or cloth in A:M.

 

 

Good ideas are never wishful thinking. There just the future in progress.

[brainmuffin]

Some info on "A Perfect Storm" :

 

http://stuffo.howstuffworks.com/perfect-storm2.htm

 

 

"The Perfect Storm" team at ILM wrote more than 30 plug-ins for Maya for this movie. They also wrote several stand-alone applications for specific aspects, such as shaders and particle systems, of the ocean scenes.

 

 

 

They didn't use particles for ALL of the water, either.

 

Basic fluid dynamic simulation provided the foundation for the body of the ocean, and became known as the bottom water in the film. The way the ocean interacts with the rigid body objects, such as the boats, allowed the team to understand exactly how each object would move in the real world under the same circumstances.

 

The bottom water was done using a deformed plane, using a fluid Dynamics simulation similar to the one Emilio Leroux (Spelling?) is looking into, I believe.

 

To make the top water (crests on the waves, the mist and foam and all the millions of water droplets in each and every splash), ILM relied heavily on particle systems.

 

And blobbies and spriticles can cover the top water effects quite nicely.

 

 

 

And as for meta balls, the only program I've seen users satisfied with the implementation of meta-balls in is z-brush. Most of the time metaball and metashape creatures are difficult to rig and animate, and it's usually only reccomended to use them if you need something reaaly quik that isn't going to move.

 

Z-brush is the exception to that rule.

[zandoriastudios]

Blobby particles ARE the same thing as the metaball object that you are wishing for.

 

What you REALLY are looking for is Fluid dynamics, like exist for some other softwares. The positions and motion of thousands of particles are computed inside a volume during a simulation, and an IMPLICIT SURFACE is calculated at the boundary of the particles. That implicit surface is what is given surface attributes of water( transparency, reflectivity, refraction) and then rendered.

[aaver]

William,

 

That's exactly what I would have liked to say

[John Bigboote]

There's a good Maya tutorial at Digital Tutors.com regarding the above mentioned deformer technique...I'll look for a link and edit it in.

[bentothemax]

Don't meta objects have the ability to melt and such though? or is that somthing else.

 

Meta objects would be good for somthing growing out of another object. Like if somone could grow another one of themselves, and it came right out of there back or somthing.

 

Or to demonstraigh cell division and such.

 

Fluid effects would be awesome though

[robcat2075]

The only problem is that they DON'T morph together like real life water particles do.

I know I'm being one of those standard forum villains by posting this but here's a render of a JohnL3D project showing blobbies morphing together

 

http://www.hash.com/forums/index.php?showt...indpost&p=15572

 

 

different settings might yield different results.

[jamagica]

You can use meta balls to make models. I once read a tuturial for a different program that has this function. I made an entire stuffed animal bear out of a set of meta balls. This is the advantage.

[brainmuffin]

I might be missing something, but I get the feeling that true metaballs would be difficult or impossible to implement with hash patches. It might be possible to get something that would appear to be the same when rendered as a metaball, similar to the way the boolean cutter works, but getting the geometry to build itself is going to be very difficult.

[NickHutson]

Exactly what I'm talking about. As to Johnl's movie. Look at the blobbies. They're bouncing like marbles, only very quickly so they appear to be melding. I've tried using blobbies for water effects and they only work for some aspects. But I feel less inclined to talk further on this line of thinking, because it would seem that no one can see or even wants to see what I'm talking about. So learning from past experiences and this thread I will try to keep my ideas to myself being that no one here is ever interested in hearing them, but only in attacking them!

 

From now on I will only post my work and my compliments on yours and your ideas. And if I don't think they'll work I'll read it again and agin until I'm sure I know what your talking about and then post suggestions to make it work if there need to be any. Thank you for your attention.

[aaver]

You can use meta balls to make models. I once read a tuturial for a different program that has this function. I made an entire stuffed animal bear out of a set of meta balls. This is the advantage.

What you are describing is an object modelling tool and that is not what A:M Blobbies were meant for. When I asked what the advantage of meta balls over Blobbies was, I referred to the water case described by Philip.

[aaver]

[...]But I feel less inclined to talk further on this line of thinking, because it would seem that no one can see or even wants to see what I'm talking about. So learning from past experiences and this thread I will try to keep my ideas to myself being that no one here is ever interested in hearing them, but only in attacking them! [...]

Philip,

 

I don't see anyone attacking you - at least I'm not and I don't think it's true that people don't want to see what you are talking about. Maybe we just don't understand what you are saying. I have read all your posts on this subject again and they still don't make any sense to me. If you think that's because I haven't understood, maybe you could try to explain again instead of saying that we don't want to understand.

 

BTW, is this what you mean when you say "morph together"?

M01.mov

[NickHutson]

Hey, sorry about that man I was in an extremely bad mood. Being attacked in the past by various other people for having ideas made me freak out this time. I guess blobbies have come leeps and bounds forward since I used them in version 4. Back then I could'nt get them to let me control them or to "FUSE", "GEL" ,"MORPH", what ever you want to call it so I stopped using it iwth later versions. But it seems to do exactly what the Metas would in the newer versions. Are blobbies what your using there?

[Roughy]

been playing with bloobies now, and ive gotto say it gives some pretty realistic water movement...well when ya run 200,000 particles atleast. unfortunatly rendering that is near...impossible

[NickHutson]

Yeah tell me about it. I ran one last night and one portin of a stream poring out of a pot took several minutes. But the results are wel worth it. especially if you use a combo of particles and diplacement maps on geometry.

 

 

 

BTW can we see what you rendered? If anything.

[Roughy]

heh, the first frame (that only has like 10 particles in it) took 16sec to render. next frame took 40sec. if it keeps following that pattern...oh god...

 

ill leave it overnight and see how far it gets....

[NickHutson]

Thanks! I can't wait to see it!

[Roughy]

behehe....been rendering for an hour now, still on the first frame ^^

 

guess 100,000 particles was a bit too much for it

 

odd thing is, theres no bloobies visible in the first frame at all...so shouldnt be anything to render, except the other crap, which should only take like 5sec to render...

[Mr. Jaqe]

odd thing is, theres no bloobies visible in the first frame at all...so shouldnt be anything to render, except the other crap, which should only take like 5sec to render...

I often sit wondering about this myself and I've come to a conclusion: A:M likes to render everything... EVERYthing. It doesn't seem to matter much if it's in front of the camera, blocked or invisible, it still has to be rendered.

 

But I don't mind...

[SHADOWMASTER]

I used them in version 4. Back then I could'nt get them to let me control them or to  "FUSE", "GEL" ,"MORPH", what ever you want to call it so I stopped using it iwth later versions.

I messed with 'em a little in v7, and they still weren't usefull for much, least not that I could figure out.

Can't wait 'till I get the new version! You're gonna be jealous, arencha.

Hehehehehe...

[Roughy]

Rofl, rendering for 10hours...still on the first frame ^^

[Mr. Jaqe]

Roughy - is multipass on? Because if not... well, you get my point

[Roughy]

well its on, but only 1 pass.

 

running for 16hours..still no change. seriously dont think this is going to finish. progressive on screen render would have finished a full frame by now, so ill have to conclude that something is seriuosly wrong here..

 

....ill stop it at 20hours or so...

[Mr. Jaqe]

You could try rendering it in preview... I don't know how much it would help, but it would deffinetly help some. After all, it's much rougher, but you'll still be able to see the result of your experiment... I think I tried doing some experimenting myself, but I didn't even consider bothering to render it. I tried, but when it said "15 hours remaining" on ONE frame, I just gave a damn. I neeed my computer, can't let a bunch of blobbies take over my computer.

 

Btw, did you have exams t'day aswell? Or are you one year younger?

[Roughy]

yups! had the written one today, guaranteed atleast a 5, maybe 6 if im lucky. Oral stuff starts next week :/

 

suppose ill stop this and try rendering low low res and see if it actually completes a single frame...

[Mr. Jaqe]

Good idea.

 

Btw, what did you guys get? We came up in math. If I don't get a 5, something's wrong, because that's the easiest math test I ever saw.

[Roughy]

bahaha, we got English ^^. ive never gotten anything below a 5 and usualy get 6 in all english tests.

 

being half american is great...^^

 

if i ended up with math or norwegian i`d prolly end up with a 4, suck at both of em

 

back on topic: well somewhere during the rendering of the 7th frame A:M vanished from my screen in the usual crash fashion. it just so happens that it would be hitting its first surface in that frame...guessing its related.

[Mr. Jaqe]

You half-english? As in England or America? 'cuz 'ere it's all American. What I'm trying to say is I'm half American.

 

if i ended up with math or norwegian i`d prolly end up with a 4, suck at both of em

Bah, easy piecy, both of 'em.

 

Btw, to get a bit ON topic, did you try to preview render it or render with a lower resolution?

[Roughy]

btw. preview render = Multipass with 1 pass.

 

rendered low with the bloobiness set to ...10 this time i think. render started at 5min, 20min at frame 6.

 

ill run it with a higher bloobiness level over night (as it really wont look like liquid without it)

[Mr. Jaqe]

Well, you have to start somewhere...