Bamf (nightcrawler Type Effect)

[chunkified]

Hey all,

I've been trying to think of how to and create a 'bamf' effect to no success

Could anyone shed some light on how i'd go about creating this or a similar effects please?

thanks

nile

[michael]

if you check out the X2 DVD there is a nice over view of the effect... Cinesite I think...

there should also be a few articles around...

http://www.cgw.com/ME2/dirmod.asp?sid=&amp...D1D2C1481651610

http://fxhome.com/forums/viewtopic.php?t=1...;highlight=bamf

http://digitalcontentproducer.com/pr/video...cts_sponsors_2/

[Allegro]

there's also: http://www.highend3d.com/boards/index.php?...212&hl=bamf

These folks had some decent success with Maya Fluids. Houdini 9 should be good for this sort of effect.

[MIguel P]

You can get nice results using noises over particles. Look at my scene, the particles have applied a VEX Anti-aliased noise vop operator and the color changes with the proximity attribute.

In the movie, Cinesite used a particle system writed by Jerry Tessendorf called "Partman" in Houdini.

Bamf2.hip

Edited April 12, 2007 by MIguel P

[chunkified]

*send Miguel a cake!*

thank you again!

[3dbeing]

OOO nice.... I like it.... I like it alot!

[sibarrick]

Very cool, so simple but so effective! good work.

[stevenong]

Great stuff Miguel!

Cheers!

steven

[MIguel P]

Thank you

But this is a very simple setup and i don't remember exactly how the bamf motion was. Of course you can do it better.

[MIguel P]

Well, after watching a reference in youtube... here is a better version with some corrections in Fusion.

If you want, I post the hip file

video: bamf

[symek]

I think particles birth should be varying across surfaces, so some prims stay alive as others turn into ashes...

just a small note on your nice work, Miguel!

cheers,

sy.

[edward]

I haven't looked at the .hip file but it doesn't seem like the object's velocity is taking into account? Maybe if the object's velocity field at the time of disappeareance could be subtly composited into the noise field?

[MIguel P]

Thanks SYmek, but in the film the effect hasn't ashes. Is just turbulent particles.

edward, the particles have in their initial velocity the velocity of the sphere, but the high drag stops the motion.

[3dbeing]

Im with edward, a little continiuation of the spheres final velocity would really unify the effect, even if it's very subtle. I think SYmek is just saying a less even distribution of the initial particle births would keep it from looking too CG.. I havn't looked at the hip either, mayber you are already doing that... And just my note, I would slow the turbulance down as the particles die, perhaps an expresion inside the popnet, where the $LIFE is a multiplier to the animated turbulance. The mostion just seem to uniform throughout the effect.

Edited April 15, 2007 by 3dbeing

[MIguel P]

Im with edward, a little continiuation of the spheres final velocity would really unify the effect, even if it's very subtle. I think SYmek is just saying a less even distribution of the initial particle births would keep it from looking too CG.. I havn't looked at the hip either, mayber you are already doing that... And just my note, I would slow the turbulance down as the particles die, perhaps an expresion inside the popnet, where the $LIFE is a multiplier to the animated turbulance. The mostion just seem to uniform throughout the effect.

Thank you. The turbulence amplitude slows down in the time, from 6 to 2, I think, maybe not enough. I'll adjust more things...

[3dbeing]

Ok so I took a look at the hip you posted and this is what I found

A: For inherit velocity to work inside the popnet you need a velocity attribute, I dropped down a trail sop to achieve this.

B: Drag should be added as one of the last nodes in the popnet.

C: Your AANoise doesn't allow for local variables IE: $ID,$LIFE, so I used a channel refrence to a null upstream to get a little variation in the particles.

Great Job otherwise, hope this helps...

-3db

Bamf3_3db.hip

[MIguel P]

Ok so I took a look at the hip you posted and this is what I found

A: For inherit velocity to work inside the popnet you need a velocity attribute, I dropped down a trail sop to achieve this.

B: Drag should be added as one of the last nodes in the popnet.

C: Your AANoise doesn't allow for local variables IE: $ID,$LIFE, so I used a channel refrence to a null upstream to get a little variation in the particles.

Great Job otherwise, hope this helps...

-3db

Thank you for your time

First of all, the scene I posted is a bit old.

In your corrected scene, the motion is not like the bamf effect. In the movie, it is much more violent.

The inherit velocity in the desintegration moment is to the left, and the drag stops the particles very fast, so I don't need too much perfection computing velocities.

The drag creates the same effect at the end of the network.

I don't want variation in the particles, because of this the roughness is 0 and the initial velocity variation is very low. This creates the nice noisy look in the effect.

Here is another version with the movement more like the movie effect: http://www.miguelperezsenent.com/Media/Tes...mf_sphere_3.avi

And I attach the hip file.

Edited April 18, 2007 by MIguel P

[TheUsualAlex]

[quote name='MIguel P

[lisux]

I could be wrong -- I believe Jerry wrote a vorticity pop to achieve a lot of these bamf effects, so it's not just some noise functions. Someone pointed out to me that POP when I was at RH couple years back... Again, i could be wrong.

But, pretty cool tho, Miguel.

I have written my own too some time ago to make effects like this, based on the whisp tool from Ivan De Wolf. maybe needs some fixing but it almost the same, putting particles in a noise field:

http://www.sidefx.com/index.php?option=com...&Itemid=148

Pretty good the effect Miguel.

[MIguel P]

I could be wrong -- I believe Jerry wrote a vorticity pop to achieve a lot of these bamf effects, so it's not just some noise functions. Someone pointed out to me that POP when I was at RH couple years back... Again, i could be wrong.

But, pretty cool tho, Miguel.

Of course, the noise method is the fast way, and of course the results are much worse than with the pop written by Jerry.

lisux, I tried with your pop, but I didn't get good results, maybe I didn't try too much... I have to try the wisp pop.