PDA

View Full Version : Advanced particle FX.



Rove
07-15-2009, 05:08 AM
Hi all,

I'm trying to find decent up-to-date (LW v9 and up) particle FX courseware.
I'm not looking for basic stuff.

I personally am highly interested in Particle FX and fluid dynamics and would like to know more about these areas. I know LW doesn't have build in fluid dynamics but it's just a thing I'm also highly interested in.

Suggestions? Tips? Commercial or not. I'm open for it.

Rove
07-15-2009, 05:48 AM
Ah I just stumbled upon Tim Dunn's commercial tutorials on the Kurv Studios website. I hope Tim will be able to take some of the mysteries involving LightWave PFX away for me.

Greetz.

prometheus
07-15-2009, 11:43 AM
Desktop images has some visual fx series with particles and dynamics..like dynamics and detail
http://desktopimages.com/VFX.shtml

Then you have some advanced particle training from asile fx
http://www.asilefx.net/catalog/images/products/lw_apd/Menu/menu.html

fluids? if you mean smoke and fire fluids? thereīs no such thing available
for lightwave unless using cantarcans dynamite fluid engine..and I donīt think theres any training on that, dynamite seems to have come to an end as a product too.

Liquid fluids..youll have to search the forums around here..thereīs some
free tutes, but that is also involving blender or real flow.

Michael

SplineGod
07-15-2009, 12:33 PM
Usually youre going to be using particles in conjunction with hypervoxels.
There are other ways to get fluid/waterfx depending upon the specific look you need.
KURV has these also:
http://www.kurvstudios.com/lightwave/lightwave3d-9-introduction_hypervoxels.php
http://www.kurvstudios.com/lightwave/water_visual_effects.php

Eagle66
07-15-2009, 06:04 PM
Something Off Topic - but i have a little problem kurvstudios.com :

There are poor information about the videos, always NO Sample Screen Shots or DVD Index and Lesson overviews or anything else what you get (LW Version?, Watch a Preview?, Project Files Included?) and many broken links on the webpage - :hey: take a look at desktopimages or asile fx.

SplineGod
07-16-2009, 01:09 AM
On the other hand KURV does provide a ton of aftermarket support via its support forums where people can get feedback and information that goes will beyond the DVD. :)

Rove
07-16-2009, 05:09 AM
Thanks for all the input! This is enough info to keep me busy for days :)

On my spare time I already have been experimenting with LW PFX. I discovered the following:

I wanted to see the difference between Particle Size on the Dynamics -> Particle tab and Particle Size on the Hypervoxels Geometry tab. And also I wanted to know if they would influence each other.
For that I made an emitter. I Specifically had it emit just one particle so I could (hopefully) see the difference better.

I found out that both Particle Size inputs indeed influence each other. I found out that I could change the transparancy of the Hypervoxel by both lowering the Particle Size on the Particle tab and increasing the Particle Size on the Hypervoxels Geometry tab. It seems to only work if I have turned on both Show Size and Output Size on the Dynamics -> Particle tab.

For now it seems it does something similar to what the Thickness channel on the Hypervoxels Shading tab does. It looks different though.

Ahhh experimenting is fun!

prometheus
07-16-2009, 12:54 PM
Something Off Topic - but i have a little problem kurvstudios.com :

There are poor information about the videos, always NO Sample Screen Shots or DVD Index and Lesson overviews or anything else what you get (LW Version?, Watch a Preview?, Project Files Included?) and many broken links on the webpage - :hey: take a look at desktopimages or asile fx.

Agreed..the marketing of and showcase of the package of tutorials are
way more attractive if you look at asile fx and desktop images, digital tutors are also very nice bundled and showcased..maybe kurv studios could
take a little note of that and see if they can improve it on their packages.

Content wise I really cant say anything about it thou..kurv studios are surely very experienced and skilled.

To me thou, it seems that many videos are to long in their length, I prefer to
watch more shorter videos and rather split up in several sections..my favourites are digital tutors..splendid distribution of sections and menus and decent lenghts of the videos..Even if they donīt have lightwave training they are good.


Michael

Rove
07-17-2009, 01:56 AM
I watched through the Kurv DVD "Introduction To Particles and Hypervoxels Vol. I".
The content is nice but the way it's brought could indeed be done better.
It was hilarious to see that Tim tried to show something which didn't work out the way it supposed to.
I knew what he did wrong though and I almost shouted at him what it was he did wrong hahaha. He finally found what caused the effect to not work right and he fixed it. All in all I learned a great deal.

dvfx
07-18-2009, 01:53 PM
Hi all,

I'm trying to find decent up-to-date (LW v9 and up) particle FX courseware.
I'm not looking for basic stuff.

I personally am highly interested in Particle FX and fluid dynamics and would like to know more about these areas. I know LW doesn't have build in fluid dynamics but it's just a thing I'm also highly interested in.

Suggestions? Tips? Commercial or not. I'm open for it.

or.....you could try Houdini. Most major Hollywood studios use Houdini for particles and fluid fx. There's a free download version called Apprentice to get you going. There are good jobs available for good Houdini artists. (no I don't work for them :) )

http://www.sidefx.com/index.php

D

Eagle66
07-18-2009, 02:04 PM
Uhh Houdini - very easy to use, learn in one week :D

On Fire
http://download.sidefx.com/images/stories/tutorials/masterclasses/masterclass_onfire.mov

Happy FX

prometheus
07-18-2009, 05:53 PM
advanced particles from tim dunne..hmm..havent seen it.

well yeah..instructors not getting
the things to work or donīt remember wich buttons or what he was doing
before could be really annoying..It might be that the video sessions
are live..and very long covering complex topics that causes that..and
it might be a little troublesome to rehearse that, but thatīs what I prefer..carefully planned tutorials or a feeling that the tutorial follows a premade script, I guess it has itīs ups and downs whatever you prefer, sometimes it can actually be useful in the sense that the instructor
misses exactly the same problem most people might run into.

I know Tim dunne is quite skilled and probably know a lot of the technology behind particles etc..and he
seems to be a swell guy and is serving the lightwave community well, and I really hope he will get going with his programming very soon and perhaps will see some fluid stuff coming someday to lightwave.

Otherwise..yeah, houdini is probably the way to go if one is interested in heavy particle,dynamics and pyro fx and fluids, unfortunatly I havent been able to try the houdini apprentice 10 version out yet due to hardware issues and network issues with my 3d computer, so I canīt connect with that one and get the license going.

Houdini is Costly in terms of learning curve and costly if you need the full master version, but for learning and students and for creating portfolios..there are cheaper versions, and the houdini apprentice program is awesome.

Michael

Mr Rid
07-18-2009, 11:45 PM
You might take a peek at these old FX Feature example scenes. They demo a lot of basic PFX and HV setups that many are not aware of.

http://www.box.net/shared/static/0jk68lcsnn.zip

Have no idea why they were dropped from recent LW contents.

DBMiller
07-20-2009, 12:41 PM
Tim Dunn's lessons are pretty good. He offers a lot of information that would be difficult to figure out on you own (at least for me). The live session recordings are a bit unstructured at times, he goes off on particle topics that don't help learn what LW can do. It's because sometimes he is talking to the students in the class but we don't see what the students are asking him.
And, yes, there are mistakes. But anything really buggy Tim will re-do on his own time and insert into the lessons. For the most part, however, the lessons are useful (as are most Kurv products) and can really help you figure out how to make the particle system work.

Rove
07-21-2009, 01:39 PM
Well the Pyro FX in Houdini sure look interesting. I guess the best thing to do is just find out about Pyro and fluid FX in general. No matter what software package.

I've seen examples where people have used LightWave Particles combined with Hypervoxels which resulted in some pretty realistic pyro FX. They claim they didn't use any 3rd party plugins. I guess it all boils down to knowledge and skills. So yea, besides playing around with what LW has to offer I'm also catchig up on some of the theory.

Mr Rid
07-21-2009, 08:31 PM
There are really two aspects to particle FX- behavior, and shading. You may get amazing particle behavior but then get stuck trying to shade them to look like anything practical. LW PFX is very basic and limited. HV is not that hard to figure if you just go thru each value and see what it does (use Viper). Age gradients are everything.

But if you are interested in physics based particles, you might try the full-featured, free demo of RealFlow. I dont think Pawel Olas has updated his Fire Smoke Generator lately. Dynamite went bye-bye. There was some other particle thing for LW that was in Japanese and probably doesnt work now either. Max and Maya are WAY better supported for particles- Thinking Particles, FumeFX, Afterburn, RealFlow Render Kit, Krakatoa, Particle Flow, and RayFire & Blastcode for shattering/demolition FX. If you want to get hired to do particle FX, learn Houdini.

prometheus
07-23-2009, 06:11 PM
The Guy above sure does some of the best fire stuff for lw particles..
and in some cases lightwave works well, but fire has a lot of different styles wether it is car bomb fire rising, flamethrower, or fire flames licking walls..or traveling
on ground or more close up flames.
Some firestuff requiring dynamic interaction are very hard to tweak in Lightwave unless using fluids.

the more dynamic and close up flames are harder to acheive in lightwave, dynamiteīs fluids was promising, but there you go..dynamite seems to have vanished now.

If Newtek had taken it under itīs wings if they could Im sure it could have
been improved to match whatīs out there today, to me it seemed to have
quality issues and missing some material options like cooling or thickness gradients, they seemed to be there but no available option to add it.
then there was no way of controlling fluids with native winds like with mayas fluids, only with objects serving as wind pushers.

Then particles couldnīt be used in conjuction with fluids either like use particles velocity and create a fluid field around that.

The opengl resolution seemed a little rough compared to maya or houdinis
open gl.

other than that, you could get some nice flames very fast and previews
thanks to viper and the fireshader was in my opinion easier and faster to tweak than maya, wich in fact could get you to some fire flames faster
than going through all the setups in maya.


I got some clips I did some while ago..merely testing it before giving up on quality, Maybe Ill put it up some day soon.

http://www.vimeo.com/user680656/videos/sort:date

Michael

Mr Rid
07-23-2009, 06:30 PM
I suspect that geometry and clothFX would yield the best looking fire. I started playing with that once, but am surprised I have not seen anyone do that in LW(?). Seems like the kind of thing Prometheus would have tried.

prometheus
07-23-2009, 07:23 PM
I suspect that geometry and clothFX would yield the best looking fire. I started playing with that once, but am surprised I have not seen anyone do that in LW(?). Seems like the kind of thing Prometheus would have tried.

ehhm..Im not really that fond of trying geometry and trying out gradients
on different channels..I believe I tried some of that quite some time ago
but I didnīt like the look of the shading..probably becaus I think it lacks
the kind of depth density and lumonosity fluid sims have, intersecting geometry is just that without adding a fuller density where geometry displacement intersection occours.

trying out cloth fx and some winds can get the movement right perhaps, and
the lumonosity and density shading look could look better if we would apply
hv sprites on that highly subdivided geometry me thinks..kinda what I experimented with on my voxel spins and plasma effect, intersecting geometry or points gathers together and increases density and lumonosity
in a better way than ordinary surface shading.

The problem with hypervoxel sprites is that when the dynamic motion or displaced meshesgets to big or strong..the points will be located to far away from eachother and thus it will give point/blob spots inbetween.
It could have been useful if we had the distance between particle parameter on a dissolve or density channel...as you are well aware of.

I sure hope someone could invent a surface shader that takes in acount
the point density and increases lumonosity based on that..or a voxel system able to work on full objects, but that would probably mean volumetrics on object ..something like creating iso surfaces from geometry
and turn to volumetric objects in houdini.

Michael

erikals
07-25-2009, 06:44 AM
...did have the cloth idea some time back, however the clothfire will look flat.

had this idea 'bout using animated displacements and a clipmap though...
(example here is not animated, but that should be fairly easy to make)
the example needs some cleanup, but done right, this should work...

(yes, i know, this example looks cartoonish http://forums.cgsociety.org/images/smilies/smile.gif)

jayroth
07-27-2009, 12:31 PM
If Newtek had taken it under itīs wings if they could Im sure it could have
been improved to match whatīs out there today, to me it seemed to have
quality issues and missing some material options like cooling or thickness gradients, they seemed to be there but no available option to add it.
then there was no way of controlling fluids with native winds like with mayas fluids, only with objects serving as wind pushers.


Yeah, we hear that alot, as you might imagine. However, you all need to realize a couple of things: if the developer has gone silent with you, its likely the same would have happened with us. In the end, the results are the same. In fact, any attempt we have made to contact this developer in the past has met with the same fate that your attempts have met.

So you see, this would have happened regardless. Personally, I am very saddened by it; clearly Dynamite had lots of potential, but apparently, not enough customers to keep the interest of the developer.

For all of you future developers out there, learn from this experience: It does not matter how bright you are, or how cool your product is, if you cannot conduct business in a professional manner.

erikals
07-27-2009, 04:40 PM
@Jay
...this guy made some pretty good fluids for LW some time back,
http://www.linkedin.com/pub/reevan-mckay/b/866/a52

screenshots,
http://www.spinquad.com/forums/showthread.php?t=8802

u might wanna contact him, maybe he is willing to help,...
might be...

erikals
07-27-2009, 05:18 PM
Hi Reevan,

I wrote to topix.com some time back and asked they could concider to release the CFD
plugin you made for Lightwave, the answer was that you had moved on to a gaming
company and they didn't know if you had any plans to release it.

So it is ok for you to release it?

I remember you saying you lost a LW dongle, but if it can help, me and some others can fix that for you.


Hope to hear from you, :)

Best reguards,
Erik and the Lightwavers

Wednesday, March 28, 2007 9:49 AM

earlier mail i wrote to him,...

found the videos... see attachments

prometheus
07-28-2009, 06:05 AM
Yeah, we hear that alot, as you might imagine. However, you all need to realize a couple of things: if the developer has gone silent with you, its likely the same would have happened with us. In the end, the results are the same. In fact, any attempt we have made to contact this developer in the past has met with the same fate that your attempts have met.

So you see, this would have happened regardless. Personally, I am very saddened by it; clearly Dynamite had lots of potential, but apparently, not enough customers to keep the interest of the developer.

For all of you future developers out there, learn from this experience: It does not matter how bright you are, or how cool your product is, if you cannot conduct business in a professional manner.

Thanks jay..that brings some lights to whatīs going on, and nice to hear you expressing your feelings about that to, yeah it had such potential.
if it really was up to lack of customers, he might have gotten more respons to it by taken care of customer support, replies and information on the page.
Personally I was in buying decisions for it as soon as I could see some more slightly improvements.

Michael

Mr Rid
07-28-2009, 10:10 AM
...In fact, any attempt we have made to contact this developer in the past has met with the same fate that your attempts have met.

Did anyone ever try contacting him in Spanish? :) Maybe he's cyberphobic.

prometheus
07-28-2009, 12:32 PM
Did anyone ever try contacting him in Spanish? :) Maybe he's cyberphobic.

Nahh..maybe this ones better to contact..

Ivé mailed him twice and got mail back each time:)

fourth of may 2009 I now asked about
if he had changed his mind or was aware of the new lightwave core coming up and if that would make a difference.
his reply..
"Hello Michael,

At this point we do not have plans for any FumeFX porting to other platform.
Been trying to find find partners, but no one seems too serious in investing time and money.

Thank you.

Regards,


Kresimir Tkalcec

Sitni Sati d.o.o.
Kutnjacki put 15
10000 Zagreb
Croatia


Michael

Mr Rid
07-28-2009, 04:26 PM
Nahh..maybe this ones better to contact...

At this point we do not have plans for any FumeFX porting to other platform.

I dont get it. XSI has ICE, Maya has its wonderful fluids, Max has particle plugins falling out of its a**. Dynamite's dead, so now with no competition in sight, here's a wide open market to all the LW users desperate for a decent pyro/smoke FX tool... arent any of these developers interested in making some money? Isnt that the point to selling anything? Should we start a petition/survey?

Lightwolf
07-28-2009, 04:59 PM
...arent any of these developers interested in making some money?
Well, the question is: Is it worth it? It's a lot of work to port something like that, especially if you don't know the target platform (that goes for both the API as well as best "work" practices to make it a useable tool).
Then counter the investment in time with the size of the LW plugin market...

Add the fact that LW 9.6 is at the end of its lifecycle (which doesn't mean it won't be used over the next three to four years, but it's discouraging).

Cheers,
Mike

Mr Rid
07-28-2009, 06:48 PM
Well, the question is: Is it worth it? It's a lot of work to port something like that, especially if you don't know the target platform (that goes for both the API as well as best "work" practices to make it a useable tool).
Then counter the investment in time with the size of the LW plugin market...

Add the fact that LW 9.6 is at the end of its lifecycle (which doesn't mean it won't be used over the next three to four years, but it's discouraging).

Cheers,
Mike

Yes, I assume there's a preference for what they know. It may be late to develope for LW. But why not before now? And what of all the current steadfast LW users that may not be going anywhere for awhile, before Core is complete and reliable with a decent particle system in how many years? How long might it take to port to an unfamiliar app? How many Dynamites are stagnating out there (and just why was it dropped)? I see a considerable market just sitting there... anyone... Bueller? Then again, maybe LW's market is that inconsequential.

So how many licenses might one project to sell of such a significant plugin that greatly extends the use of an animation app? I know a company won a considerable bid on a batch of feature film FX shots primarily because they were able to prove they could do convincing smoke FX in a test, using a last minute acquired FumeFX. This is the kind of investment that can distinctly influence job opportunities.

Apparently it has been worth supporting LW for NextLimit and... umm... what other equivalent plugins designed to support multiple 3D apps currently include LW? I mean something like Endorphin, Thinking Particles, Craft Tools, Shave & Haircut... all I can think of offhand is RealFlow, although NextLimit's recent RenderKit did not bother. I could list many 3D services I have run into that dont support LW. And why is that?

I remember the last hour at Siggraph '06 being told by Natural Motion that I was the only LW user to enquire about Endorphin (what I thought was the coolest product there)... in all of Siggraph... Back when I was messing in RealFlow 1.3 for about a year, and emailing support every other day (best support I ever dealt with- always responsive and patched the app regularly per queries/comments), I learned I was the only LW user that NextLimit was hearing from. I keep wondering, what the hell are the rest of you LW geeks doing?! :beerchug::sleeping::hammer:

erikals
07-28-2009, 11:20 PM
...clearly Dynamite had lots of potential, but apparently, not enough customers to keep the interest of the developer...

well, this is not the reason. any Lightwaver that actually bought and used the plugin will state that the plugin is pretty much useless, making artifacts ("blocks") in simulations. so, clearly this has nothing to do with us, but the developer that didn't finish the product, in fact, if it had a higher price, i would claim my money back, but i just won't bother.


I could list many 3D services I have run into that dont support LW. And why is that?
because most companies having the capacity to make major VFX is using Maya, and of course these are the companies that have the ca$h to buy them.


I keep wondering, what the hell are the rest of you LW geeks doing?!
well, personally i'm just a freelancer, so i don't have the capacity. the only app i can think of using that i could benefit from is Realflow, but even that takes time to learn and simulate. and i cannot guarantee that the outcome will be satisfactory within the timeframe, as RF simulations take so long to compute.

my guess, Lightwavers do not have the time available / capasity to make heavy VFX.
take Battlestar Galactica for example, basically all explosions are done in space, which is just a matter of pasting some footage of explosions onto the renders (ok, put a bit simply, but u get it)

and why would any waver at a VFX company ask for smokefluids when the crash site using LW RBD looks like crap?

no Sir,... no smoke without fire!

erikals
07-28-2009, 11:45 PM
...however, like stated in another thread, there is a free app out there that does very good RBD, Cloth, and soon,... Smoke...

http://forums.cgsociety.org/images/smilies/wink.gif

prometheus
07-29-2009, 12:16 PM
Ivé just put up five more little short clips of some particle test..the god old
try smoke and flame tests, one dynamite fluids I did over at a friend some time ago and the rest is plain ol fine tuned particles, it might be useful to render out and map on sprites or geometry even, the particle amount is fairly low for this kind of stuff me thinks..around 6000-20000 particles.

http://vimeo.com/user680656/videos/sort:date


Well..I hope that Tim (Aurora) dunne will be able to come up with what he calls oto taca..
check the thread here where he mentions it.
http://www.spinquad.com/forums/archive/index.php?t-25240.html

As I understood it, He couldnīt really get going until some things in core was finished or something.

Tim seems to know a bit about fluids, and if Im not mistaken he is working on a bachelor degree or something.

If we were to push it a little further instead of just catching up with the
ol fire and smoke fluids or liquid fluids, I would love to se a hybrid of fluids and the popular chaoscope fractal generators/attractors or like the Auroras plugin for lightwave or the spore program that was used in solaris.

I tried the aurora plugin,cool stuff..but it didnīt reach the level of awesomeness I would love to see.
chaoscope are pretty darn sweet..would be nice if it was fully 3d thou.

Im no tech guy so take this with a grind of fluids, this is just me dreaming.

One thing about these attractors is that they yield some cool stuff sure..but I would like to have a system were you can create attractors anywhere in 3d space by simple mouse clicks or using a tablet or using ordered structures from L-systems or point geometry.

these attractors could have different powers at the same time, lets say you place 10-30 attractors anywhere and you can decide what type they should be by default, like attract..if we were to have a couple of attractors somewhere..the particle stream would start moving to
those attractors, thatīs okey..

but if we at the same time just select any of these attractors..or if we have a tab menu option open we can adjust parameters directly and activate other forces like vortex,rotations,spins etc..and this could be controlled at one attractor at the time or at several attractors on the same time.

the particles streams should have different spreading options or conserving
velocity,energy,mass etc.

These attractors should be fully able to move by driving the motion by
premade particle systems motions,motion capture or geometry displacements,fluids or even sound,anything goes..
isnīt that what core aims for:)

The combined motion of the attractors and the following streams of particles or fluids could create really stunning effects and itīs a lot
more control over the looks than those fractal generators but still
with undulations thanks to different attractor forces that are multilayered.

The option to choose to render it out with a particle render ala krakatoa..or generate fluid voxels from that should of course be there.
maybe Ill have to do some sketches, blueprints of what I mean how this could look like.

I guess this would take some serious time to work on and very complex calculations,
but hey..aim for the stars.

oh..I forgot...donīt forget the distance between particle parameters Newtek:)

Michael

erikals
07-29-2009, 04:26 PM
oh..I forgot...donīt forget the distance between particle parameters Newtek:)

just wondering, ...is there any way to achieve this now somehow?

Mr Rid
07-29-2009, 10:40 PM
just wondering, ...is there any way to achieve this now somehow?

*sigh* :cry: I really cant think of a simpler update that would go so far to changing what you can do with a LW tool. The data is there, NT just wont deal with it. I wish one of these L-script, plugin-minded, techy guys would whip up a little thing for this. That useless HV deform ought to address it.

A number of 'features' have been thrown into LW incompleted. There are Distance Between Particle grads in some HV channels, but that dont actually work in most of them, and are not in the channels where they are most needed- blend, and size. Even if they did work, it can get tricky stacking grads for all axis. There needs to be a global option.

Here's a scene illustrating the problem (the top 4 examples dont work) if you make a Viper scene preview. What's weird is they work in GL but not in the render(?). Yes, it was fogbugged long ago. Thew ones that do work are snapping too quickly.

75812

http://www.box.net/shared/static/m1sqicrwo4.zip

prometheus
07-30-2009, 01:17 AM
*sigh* :cry: I really cant think of a simpler update that would go so far to changing what you can do with a LW tool. The data is there, NT just wont deal with it. I wish one of these L-script, plugin-minded, techy guys would whip up a little thing for this. That useless HV deform ought to address it.

Agreed, it would be sweet if dpont could whip up something with his nodes.
Not pretending to be an expert on this, but this implementation would probably have a huge impact on particle effectīs realism.
Donīt do tomorrow what you can do today.

Michael