Particle Flow Discussion

#141

Nice job sidvici. Its looking really nice. :slight_smile:

#142

I would be posting more, but I have been swamped to an extent that I have not seen till now at work. Let me get some of this work that I have in front of me out the door and I will get back into this hard and heavy.

Thanks for the enthusiasm.
:bounce: :bounce: :bounce:

#143

sidvici and BigRanS123, I solved my glu3d problem. I just deleted the 3dsmax.ini config settings and restarted max. :slight_smile: :slight_smile:

#144

how did you hit on that? thx by the way.

#145

Ih
I m trying to approach to a good flame motion with PF

the render is quite fast : 6second/image 320x240 (dual-1.5Ghz)

Tell me what do u think of my tests
thx

Loran

[http://forumel.free.fr/flame01.mov]

#146

the turbulence looks cool. some kind of plasma fire. you could adjust your material and particlesize. looks a bit blurry did you use glow here?. maybe you can upload the scene.

#147

Cool fire! Probably not applicable in this situation, although with fire making the material additive makes a big difference. If you adjust the filter colour to be brighter, it’ll intensify based on how many particles layer on top of eachother, which gives a nice blown out look to it which usually for fire works pretty well.
Cool work!

#148

Did some firetesting today, and thought I’d make a ball that drops down some stairs. At first it looked kinda cool as it was - then some guys on IRC started complaining that the fireball didnt leave any marks, so I hade to make some marks on the flor as it rolled down. So I spawn 5k particles every time the ball bounces down the stairs. Combined with 300k particles for the flame, you get a pretty cool effect and one hell of a rendertime :slight_smile:

http://frodo.hiof.no/~deetee/files/3dsmax/fireballroll.wmv

#149

Wow, thats really, really nice looking fire deetee. That dosen’t look like 300k particles though.

#150

Actually I lied… hehe. I had a little type-o, it was more like 200k particles and not 300 :slight_smile: The smaller the size and higher the count, the better the flames :slight_smile: I actually started out with 1mill particles with scale 1, and end up at 200k size 3, cause of the deadly doses of information I fed to my CPU…

http://frodo.hiof.no/~deetee/pfupload/?act=view&file=YmFsbGZhbGwubWF4

There is the .max files - rip it apart boys and girls :slight_smile:

#151

Sorry i’m late but i have been really busy latley, but here are the .max file for the my particle flow fire.
The flames are more based on procedual materials than actually using pflow to create realistic fire-motion.

MAX : http://frodo.hiof.no/~deetee/pfupload/index.php?act=view&file=dGhlb19mbGFtZWluZ2JhbGwubWF4

PICTURE : http://frodo.hiof.no/~deetee/pfupload/index.php?act=view&file=dGhlb19mbGFtaW5nYmFsbC5qcGc=

AVI : http://frodo.hiof.no/~deetee/pfupload/index.php?act=view&file=dGhlb19mbGFtaW5nYmFsbF9yb2xsaW5nLmF2aQ==

I know the avi and the .max file are not the same , i lost the original rolling ball max file but atleast the particle system is the same :slight_smile:

#152

WOW Theo, at first i thought this is an easy ab- render. but you did it very well without ab. man, that smoke looks damn good. i tried something similar with mblur spheres but this is a much better solution. very nice!

#153

sidvici : thanks, i know there is some small issues with this kinda setup, since i’m using shapefacing you do get certain limitations as to how the fire/smoke moves relative to the camera and such.
Another thing is that some of the fire parts are not blending into smoke but gets deleted. I think it is because the particles do not get enough time to blend and thereby causing them to move to an delete operator. I would really like to try and make a really complex smoke/fire setup that uses multiple pflow systems and 3-4 main materials with all animated phasing , but then again :slight_smile:

Btw, great thread! Keep it up!

-theo

#154

loran:
I like the fire that you ended up with, but I do agree that your particle size needs to be smaller and have the color of the flame fall off some near the end.

amckay:
Hey man thank for stopping by, I have learn the majority of what I know about p-flow form you site. thanks.

deetee:
Wow, I like the look you have here. I under stand what you mean, the computers don’t like you when you start getting into the Millions of particles.

theotheo:
I know what you are talking about. I have one or two files that I am working on that has many different p-flow layers to it. They all either look bad, or take to long for me to render right now. One day…

I should be getting Combustion3 in soon, and I plan on rendering some smoke and fire out of there and trying to use them as sprites like thy did in LOTR on the balrock(sp). I hope that I can get some cool effects form this. Does any body have a good explanation for how they did the fire in that particular scene, besides the DVD extras?

I hope to get some tests up here soon, but as for now I’m swamped, and that’s a good thing, right??

#155

yea, rans i know what you meen about being swampped. ive got a lot to do at work and im working on some stuff on the side in every bit of free time i get.:thumbsup:

it seems that everyone has a pretty good grasp on the creation of fire and the different tweeks for different effects, but has anyone tried a NUCLEAR EXPLOSION? that would be awsome with the pflow system. i am intregued and if i can round up some free time i think i will begin to try. NUKES ROCK:buttrock: :buttrock:

#156

I don’t know why but NUKE explosions have always troubled me. That would be nice if I could set one up that looks decent though. :slight_smile:

#157

doing a nuke can be a pain the ass.
i think a lot of the time doing stuff with geometry and other effects can sometimes be the most effetive way to tackle that kind of stuff, as doing it with particles tends to require a lot of work.

Using fluids can achieve some really nice stuff, like what they did with t3 using a 2D fluid vector field to control a spliced 3d fluid simulation… Ron Fedkew at Stanford helped R&D all that stuff for ILM. It’s a cheap and simpleway to approach the desired look you’re after without having to go through gigantic simulation times when designing the look.

#158

Ok cool Allan, thanks for the info. I see what you mean by using 2D fluid fields to get the desired look. Particles would take a lot longer to setup and much longer rendering times too I bet.

#159

well simulating a nuke with particles is a lot of work. ack, I actually did a vid tut for my dvd which covered this but it was recorded on a botched capture card so I turfed it. But essentially if you wanted to get that kind of motion you’d probably get a lot better results with fluids, at least a lot easier.

Personally if I could get away with it I’d never really do the nuke itself, just the aftermath… big lenseflare blow out, and then the head with a bit of motion, and focus on the destruction that it causes.

Look at sum of all fears, R&H pulled off some great stuff without having to do any actual rolling big nuke, just the mushroom cloud after it’s formed.

#160

Too bad max dosen’t have a good fluid system yet. Theres probably ways of faking it, I just don’t know how. Ohh btw, I like your drip system method on your site. Could you tell me how you got the water to follow down the sphere? That was really neat. :slight_smile: