Nuke Explosions..for Ka-boom freaks only!

#121

Thats not my plugin, I believe it creates new voxels based on density hitting the edge of the container. Downside with it is you cant cache a non-constant sized container so a little worthless in production, but a nice tool nevertheless.

#122 
 I used 2 fluid containers for the ground and mashroom parts. The ground part has 160 40 160 resolution and its density buoyancy is  -1. It has an animated emitter with a denstity of 50 and a heat of 20 for 2 frames.  This fluid is driven by volumeAxisField with a mag of 200. The mashroom part  is 100 160 100. It has an animated density/heat/fuel emitter for 2 frames. The fluidShape settings are in the pic in the attach. The debris particles affect the voxel velocity they are going through (expression).

 Thanx guys for your comments.
 
 Best regards,
 Vladimir
#123

I was thinking that this could be the best plug-in for me now but…T_T.For what I’m trying to achive I need rly rly tall fluid container,but becouse I’m using animated fields I will have to reanimate them if I make even a little.When I try to reanimate them it looks like that the fields are not working on the fluid anymore T_T O_O,maybe becouse I del the keys from the time slider I dont know.Anyway this is what I was trying to get close to this week http://www.youtube.com/watch?v=hfEMnx-Nz-w (not the shape the rolling :> ).And this is as much as i can go http://www.youtube.com/watch?v=n6GVREqRwxQ.Its kinde a rolling as I want it to,but I dont know if it continuing and for how long :>.BTW boyancy here is 1,but when I try it with -1 its almost the same results xD.

I used 2 fluid containers for the ground and mashroom parts. The ground part has 160 40 160 resolution and its density buoyancy is -1. It has an animated emitter with a denstity of 50 and a heat of 20 for 2 frames. This fluid is driven by volumeAxisField with a mag of 200. The mashroom part is 100 160 100. It has an animated density/heat/fuel emitter for 2 frames. The fluidShape settings are in the pic in the attach. The debris particles affect the voxel velocity they are going through (expression).

I was wondering if you guys was using another plane for the ground part thx for the share :>.

#124

Hi all,
Just wanna share my little conversation with Duncan on Maya Fluid behaviour. I felt this might help others as well in carrying their test with maya fluid So I am putting up here:

Hi Vik,

Both swirl and high detail can result in instability, but high detail is more likely to. (the normal fluid solve without swirl has been mathematically shown to never blow up) Instability generally results when using high detail where the velocity divergence is high relative to the voxel. Thus when you have higher resolution the fluid may be attempting to move several voxels in a step which can lead to a blowup. The velocity relative to voxel size increases with resolution, which makes high resolution more unstable. You can avoid the blowup by caching with “evaluate every” set to 0.5 or lower. It will then take more steps with proportionately lower velocity per step.

Another thing that can help with blowups is a very small amount of damp on the fluid. As well one can avoid things like making the temperature turbulence too high( or anything else that generates high velocity ). Fast moving fluids (with high detail) require proportionately more steps to solve than slow moving ones.

I would ultimately recommend high detail solve on all grids for the best result(caching on subframes if needed for stability). Avoid using swirl with high detail solve as one gets too many artifacts and it makes it less stable. If you don’t use high detail solve, swirl (which is the technique of vorticity confinement, if you want to search the web) is pretty important for detail, but it only creates small scale vortices and can look a bit artificial.

To keep vortices moving fuel can be useful. As it burns and releases heat this can drive a rolling vortex, much like hot water in the gulf can feed a hurricane. Just emitting temperature will diffuse and heat too many areas, while with fuel one can concentrate the added heat inside the vortex.
Temperature turbulence is also critical for a good effect. It is a bit tricky adjusting the burn rates and heat released to get the right balance.

Duncan

Hope you all find this very useful in doing your Sims.
Happy Blasting,
Vik

#125

Thanks for sharing Vik, and I must thank Duncan for the great tips. This is extremely helpful. I normally use swirl/volume axis fields to drive movement for most of my tests. I will certainly sim some tests without swirl, driven by flue to see the results. Doesn’t hurt to have alternate solutions.

#126

Awesome and helpful this show me that I was at wrong direction(I used a lot of swirl like 25-30 for my last sims,but I usualy use 5-15 O_O :> ).
Thx for sharing man :>.

#127

For the ground bit, Ive used an animated nurbs circle curve as an emitter. Scaled it up during animation and then dropped it out the bottom of the container. PRoduces a nice effect even if its a bit too even but it means you dont have to worry about another container or using flieds for the ground blast bit.

#128

I’m trying to make that ground boom now :>.Thx for the advice man :>.

#129

hey all, just wanted to let you know that this thread is a WEALTH of information for the new maya user interested in fluids. Thanks to all participants thus far! I have a question one of you may be able to answer. I’ve got a decent fluid sim going that I’m happy with, but I’d love to have it run at about 1/10th the speed it’s currently running…if I try this by time scaling the cache, it gives me stepping and ghosting (not “smart” interpolation) and if I try to change my simulation rate scale, my smoke quickly gets stripped off most density and I’m left with a puny ghost of a smoke cloud. Any thoughts on the matter would be greatly appreciated.

thanks again guys!

*edit - thought I had it…not quite :frowning:

#130

One test rend(still working on it :> )
link

#131

Great thread going here,

Read it all, but think im missing something, not really getting close enough.
I’m stuck at the moment, to much possible settings.

Color, incand and transp have a inflected billow texture.
i used no fuel.

Sim rate 1.55, high detail solve, except velocity.

damp o.o8
density boancy 0.5
temp scale 1.0
temp boancy 50.0

turbulence 10

emitter,

dens voxel keyd from 60 to 0 ,frame 30
heat 15 to 0.1 at frame 15

Now i m stuck with the settings, want to get it to go slower, and look better offcource.
been playing with it for a day now.

Any suggestions?

#132

If you increase the resolution and lower the simulation rate scale then the simulation will tend to behave the same but more slowly and with better detail.

#133

hey Folks,

Just to give you guys heads up, here`s a little test scene file for all nuke freaks. Just sharing my little research on maya fluids. This test shows about the rolling behaviour in maya fluid.

Many of us believed that rolling happens coz of swirl which is not correct It happens coz of heat. (Duncan stated that before coz Of which I was able to justify my test).

I think swirl is only a multiplier for rolling when it comes to nuke explosions. It helps to add extra detail or a noise. It adds some kind off turbulent noise or may be a small vorticity to fluid motion.

I always visualise my sim with velcoity vectors as it helps to see where & how ur fluid motion is leading.

The key for continuous billowing/rolling effect in fluids is what I found is heat & thats what my test shows.

I think the best way to make rolling fluid is to use high amount of temperature or fuel which has high heat release value & one can control the color shifting through reaction speed instead of keying incandenscence ramp.

I think it’s more important to treat the air around density, than the density itself to cause rolling. So I ran a test without emitting any density just temp/heat & visualised the motion of velocity vectors & they were rolling as expected even if the swirl value was set to zero.

Some quick recipes of how I cook my explosions in maya fluid container :slight_smile:

  1. Make Density scale higher than 5. A good/crazy amount of density used in conjuction
    with heat can give more explosive results.

  2. Emit loads of density for a few frames & then key off to get the puff imapct in the start.

  3. Heat : controls “explosiveness” of reaction. It also responsible for making fluid roll :).
    heat can be used best in conjuction with fluid density emission & fuel emission.
    High numbers create combustion effects like explosions & fireballs, Overall effect is
    faster & more aggressive.
    Low numbers create effect like candles, camfires, things burning. Overall effect is calmer.

for fire : If you want to increase ur flames speed over the time increasing temp turbulence
& temp buoyancy will give u much better result instead of increasing simualtion
scale rate.

  1. I normally do my tests with “High Detail Solve - all grids except velocity” way, which is
    quite similar to All Grids except velocity is unaffected by High Detail Solve.
    this is useful to retain most of the benefits of the extra fluid cohesiveness without much
    of a higher calcualtion time.

  2. In builty Gravity : do not get confused by the name.increasing gravity wont make your
    fluid go dowm instead it will make ur fluid more light in space.
    I often use it make my smoke move faster in wtever direction they are been pushed.
    Its a good way of making ur smoke reacts fast with field`s force instead of using Density
    buoyancy as density buoyancy will make fluid go faster in Y direction only.
    Gravity has indirect relationship with air So for e.g If have to make fluid thicker or give
    more mass so I will reduce the gravity value.

  3. Fuel : specifically controls amount of flammable material given to simulator.
    i.e Lownumbers mimic kindling (possible low quantity & less flammable),
    high numbers mimic gasoline (highly flammable). It specifically controls speed of
    reaction depending on what values you keep on speed reaction & ignition temperature.

  4. Using Fields always helps in giving ur simulaiton a GOOD SHAPE. I highly prefer using
    volume axis field which gives you good control to your sim, of how fast you make
    your explosion grow & expand over time.

Above all the most [color=red]two Important*[/color] thing you need is One Beer & a good Progressive Music :slight_smile:
I couldn`t have done all these without the most important thing. they will keep you in rhytm all the time :slight_smile: lol! :thumbsup:

hope to see more maya fluid on this FXWAR show.
Cheers & happy blasting on this valentine day!
Best,
Vik
http://www.youtube.com/user/viki164
http://forums.cgsociety.org/showthread.php?p=6357279#post6357279

#134

these the above scene file image :
As u can see my velocity vectors of how they behave here to give me a continuous
rolling/billowing effect without using any swirl value. hope the idea is much more clear
now.


Best,
Vik

#135

Vikas this is great.
Really great. I was in a moment of study to make the nuke shape but i couldn’t afford to lose a lot of time to it, so I was thinking I would’t be able to finish it on time.
This is huge data for me.
Thanks a lot really.
Hope I can pay you this act of yours.
Good luck to you.
Cheers

#136

@VTeixeria :

Hope I can pay you this act of yours.

No need brother :slight_smile: , I dont need any thing. may be you can help some animal welfare or any organisation of what u feel good.

I really appreciate ur enthusiasm & dedication towards Fluid study. your fluid group videos on vimeo is very helpful & also a very good place for references.
your efforts reminds me of my earlier stage when I was new to Maya Fluid.

keep up the good Work,
Best
Vik

#137

Check out these links if you are interested in knowing a bit more whats behind the fluid simulation softwares.
This is my research for my Master thesis.
At the moment I’m devoting to RF I’m finishing the water simulations but I would really liked to achieve some nice results in this challenge, just for fun.

http://www.realflowforum.com/view_topic.php?pid=30108

http://forums.cgsociety.org/showthr...p?f=86&t=838205

http://forums.cgsociety.org/showthr…?f=183&t=839382

Hope this information is useful and interesting to start a good discussion.
Cheers

#138

Great tips Vik! I’m definitely going to try these techniques. Curious to see the results with the settings you mentioned.

#139

Thats an epic fluid Sim Viki164!

This is a pretty epic thread in general! Ive been working my way through a lot of the pages and
have been experimenting - trying to get my head around fluid explosions. Without finding this thread I think I would have given up on maya fluid explosions ages ago.

My basic tests are on Vimeo:

(they are a bit dull after some of the recent posts but thought I’d share them anyway)

http://www.vimeo.com/9449173

Any suggestions for improvement welcome. I’d really like the blast to feel much larger but find if I lower the sim scale it loses a lot of drama/dynamic shape

There are several settings I’m not sure about:

Solver Quality: I tend to stick it on 100 but cant see much difference if I go 50 either way. Is this an important feature I need to be more aware of??

How can I get more detail in to the fluid? Adding more temperature turbulence makes it noisy, and up close the fluid gets quite voxelly(kind of square blobs of detail). Do you think I need to increase the resolution or start using textures? I’d seen several posts saying avoid the texture option so haven’t touched it yet.

I’m gonna try some sims without swirl after reading the most recent posts.

Keep up the good work. This thread is a wealth of info & ideas

cheers
Neil

#140

@ Nwilco :
I didn`t say not to use swirls. I was just justifying my test that swirl isnt responsible for rolling its heat. Like I said swirl adds extra detail noise.
I normally dont keep swirl more than 15 or else it might give more noisy artifacts.
so feel free to use it.
btw your test are pretty neat & I think you are getting there.
best
Vik