Nuke Explosions..for Ka-boom freaks only!

VTeixeira · 2010-02-19 07:49:43 UTC

Your book Todd gives me some very nice tips also.
It’s been helping me a lot.
I’m still fighting with that ceiling animation but thats in another thread.
Thanks

Very nice tips Viki but unfortunately I’m not in that level at the moment.
But I’ll reach it :bounce:

Ps: I’ve tried to take a look at your website but it’s not working.

Skittixch · 2010-02-19 12:28:41 UTC

hey all, just thought I’d offer some unsolicited advice.

I was having a hell of a time getting enough “edge detail” out of my simulations, and I finally broke down and did some high resolution 2d fluid sims with a similar trajectory, and overlayed/manipulated them in my compositor. It’s really upping the quality of my shots.

Keep it up guys!

VTeixeira · 2010-02-19 12:52:29 UTC

Could you show some examples please?
Cheers

Skittixch · 2010-02-19 13:34:25 UTC

I would like to, but for now, these shots are slated to air on a network tv show sometime in the next few months, so I can’t show much of what I’m working on…but I’d be glad to give an example of the workflow at some point soon. I’ll keep in touch.

btw…a friend of John Cage is a friend of mine

fxabhi · 2010-02-20 12:16:59 UTC

@ viki … its an ma file… what? i have got an mb file… is your file named nuke5? cause that is what i got when i downloaded the attached zip? that’s the very reason i didnt try any stunt’s with that mb file like notepad and stuff. ill try downloading again i guess…thanks

edit… i am still getting an mb file…am i doped or is it really an mb file there… people can you tell me if you guys are also getting an ma or an mb file which viki shared recently named nuke5.zip ( i have seriously lost it… to be asking this)… is an ma or mb file… and viki if its not too much to ask can you post it again …

Cheesestraws · 2010-02-20 12:24:36 UTC

Just turn on “ignore file version” in the load settings.

fxabhi · 2010-02-20 12:32:44 UTC

@cheesestraws … “ignore file version” that option has been like that for a long time now on my machine… i always keep it on… still nothing… thanks though.

VTeixeira · 2010-02-20 12:50:38 UTC

The file was ok to me.
Great one should I said.

btw…a friend of John Cage is a friend of mine

You are the man
Cheers

fxabhi · 2010-02-20 13:03:54 UTC

this is freaky… the first 5 times it never opened … i downloaded this thing again … it didnt open… i restart my machine and bham it opens… sorry for being a 5 year old fusspot making a hula hoo about it… thanks VTeixeira and Cheesestraws for the support at 2.30 in the night here in india … lol…now that the file finally opened ill see it in the morning … why is me so lazy i have never understood… the file is working and i want to sleep…any ways good night and thank you viki for the file, great file

Diependaal · 2010-02-23 14:05:23 UTC

tokanohanna:

Hey Neil,

      To add to the great tips from Vik and Duncan, I would suggest maybe turning down your solver quality.  100 is probably over kill. Yes a high solver quality will increase the accuracy of your sim, it pushes simulation time through the roof.  So try and find a good medium for your container.   Be very careful with using the temperature turbulence, it does a nice job of giving detail to the fluid, but like you discovered.....too much will cause noisy artifacts.
      
      I took a look at Viks scene file and, I think driving the fluids with temperature and fuel/heat is great, it certainly allows more control of the billowing shape of the fluid.  And it's a great way and an alternative to texture your fluid instead of keying the incandescence input bias.  The disadvantages of using this method is, calculation time is increased depending on your system.  The fluid also may lack irregularity within its shape, causing it to look unnatural.  All explosions drift and have distorted shapes.  It never just goes straight up.  I have been guilty of that in early tests that I have done.
      
      I did a quick test to compare fluid driven just by swirl and buoyancy to fluid driven by heat/fuel.   [http://www.vimeo.com/9459919](http://www.vimeo.com/9459919)
      
      No temperature/fuel.  Just Density and Velocity is used under contents method.  The fluid emitter also is emitting 0 heat and fuel.  I found that the fluid still maintains a nice billowing shape.  But it lacks the explosive force temperature/heat can provide.  And it also loses it's shape at the end of the sim.  You can easily make up for this by increasing your resolution so the fluid will maintain detail.  The key to utilizing swirl with high detail solve is making sure, you key your swirl parameter to a low amount after the fluid takes a nice shape.  If you leave it high throughout, your fluid sim will break on you.  I have swirl keyed at 10 to 1 from frames 1 -48.  To make up for the force, just add a volume axis field to create the explosion.  The volume axis field basically is fuel for your fluid. To fake the look of a fuel driven container with high Temperature scale, you can cheat a bit by making Temperature dynamic under Content methods keeping Fuel off.  Keep your temperature scale at a moderately low value(.5-1),  giving your emitter a very small amount of heat(.5) leaving fuel at 0, and simply keying your incandescence to shade your fluid to your liking.  
      
      I would be careful utilizing a high amounts of the density scale parameter because you want to maintain a cohesive relationship to your transparency and your opacity input bias parameters.  Density higher than 1.0                   or greater could actually be quite sudden and unnatural.  All smoke has some type of transparency.  I would probably recommend tweaking the density/voxel /sec parameter of the fluid container instead.  
      
      Notice between the two tests the advantages and disadvantages I mentioned above.  Overall I would recommend using heat/fuel to drive your explosion.  If you find a hard time creating irregular shapes of your nuke then I would suggest creating and animating a few volume axis fields or motion fields to break up the billowing movement a bit.  
       
      Basically, you can achieve a great explosion using both methods.  Just depends on your shot, how much time you have, and the system you are using.  So don't be afraid to try both or combine the two.

fluid driven just by swirl and buoyancy compared to fluid driven by heat/fuel.

For the last you still need buoyancy right? Because cant seem to drive it up any other way. accept for maybe freaking with the velocity scale.

kdronez · 2010-02-23 14:25:29 UTC

You can use only temperature buoyancy to drive the fluid :>.

kdronez · 2010-02-24 06:57:49 UTC

Here are some more tests using animated velocity scale.
I cant get a tail becouse of the velocity scale,but "the cap"looks ok T_T.Is there a plug-in or expresion or smth that can control the velocity in different parts of the container?If not pls put feature like that in maya 2011 hahaahahha xD.
I’m thinking to make new container for the tail :>,but first I need to make more tests so that I can se the best values for the tail xD.
test

fxabhi · 2010-02-25 12:41:55 UTC

ok am back to make a fool out of myself again … but this time … i have done some tests… now after getting vik’s file to open in maya … that morning i decided not to look at it… ahem… (as if anyone would believe me)… but thats what i told myself … atleast once i should try to get somewhere close to something which looks like a rolling mushroom…so here are the tests …

http://www.youtube.com/watch?v=3qgfN5KsGqI

slower video… http://www.youtube.com/watch?v=b4rG2RuyElE

i think the last one looks … something i can say close… but i like vik’s video results much better … so now i think its time to earn vik’s file … its like having bournville … you dont take it you have to earn it… lol… i have nuked my brains out…

damn that mushroom… thanks vik for the file … i hope after looking at your file i can get better results…

need your crits please…slap it!

cheers…

VTeixeira · 2010-02-25 13:07:09 UTC

Well I really liked them.
Definatelly you are in the right way.
Congrats

fxabhi · 2010-03-01 01:14:38 UTC

Thanks VTeixeira well here is a bigger version of the last blast with different emitters though…

http://www.youtube.com/watch?v=GycYTkH6gps

i was thinking of combining what i have understood till now with this blast i did 2 months back… i guess this should … bring me somewhere… http://www.youtube.com/watch?v=02GOCtPKt10&feature=related

although i have to work on my timing a lot … my timing sucks here…:argh: any suggestions for the timing of the implode and the color shift…

and VTeixeira i just saw this video on your gallery i think… this guy had made it in 3ds max … using krakatoa… it was crazy with music and stuff… i also tried my bit lol… CC more than appreciated…

check it out… http://www.youtube.com/watch?v=q1nMLPmp7-A

cheers and thanks again…

VTeixeira · 2010-03-01 04:45:15 UTC

I think it’s great.
Really.

fxabhi · 2010-03-01 21:15:14 UTC

yep VTeixeira, i believe ya… hahaha…

kdronez · 2010-03-03 13:57:00 UTC

Fxabhi its rly nice I like it too man.
Anyway I manage to make sooooooome tail on my velocity scale method that I’m testing,but its still no good and it needs lots of work :>.
test continues xD
Probably I will stop for a while,becouse as you probably know SC 2 beta is out aaaand I’m waiting for this game for almost 10 years now sooooooo you know its time to play a little xD.
PowerOverwhelming xD ,/ ^^.

fxabhi · 2010-03-03 23:15:44 UTC

hi kdronez, the left one looks cool… what res do you have on both…. edit " oh you have mentioned … nice man for a 110 res… "

keep them coming with time… we guys will nail it :p…

cheers

destruct007 · 2010-03-04 13:26:32 UTC

2 weeks late, but in honor of Vik’s Birthday I did a very “Vikas” test. A Kaboom in playblast form ;D

http://vimeo.com/9918616

actually I was inspired to mess around with Fuel more after reading Duncan’s response to the fuel question thread.