Hey

I' am trying to rebuilding a job I did 2 years ago, which was heavy on the particles, rbd and fluids. It was done all in Softimage.
http://www.sekow.com/catrice_color [sekow.com]

Seems to be the best candidate to rebuild it all in Houdini.
As every effect had something not trivial to it, I had to use a lot of ICE trickery, which I want to adopt in Houdini.

Starting with the dust explosions. In ICE I've grouped a big load of particles into separate clusters, gave them the same velocity. after a designated timeframe the clusters started to fall apart leaving this beautiful trails behind them.
Achieved through some array fumbling and filtering… well in general.

Indeed, this is a great project to create in Houdini.
I dare say you will find some shortcuts when compared to SI´s approach.

well for the moment i am struggling a lot.
not having the whole math/logic in one graph, and separating creation from simulation, is very (VERY!) frustrating.
I blame it on my non existing experience and misconception, but I thought it would be easier, knowing what I need to do technically…

I will give it some more try, before I start spamming this thread with questions.

wow it is looking nice….. i like it. :-)

it's depending where do you struggle?
there are alot of tutorials with plenty of informations.

don't forget look at the masterclass tutorials:
http://www.sidefx.com/index.php?option=com_content&task=blogcategory&id=134&Itemid=344 [sidefx.com]

for the bullet solver i like this one special:
http://www.sidefx.com/index.php?option=com_content&task=view&id=2350&Itemid=344 [sidefx.com]

from peter quin there are also some tutorials to point out:
http://vimeo.com/user2030228/videos/page:1/sort:date [vimeo.com]

Don´t worry…spam away!

sekow
well for the moment i am struggling a lot.
not having the whole math/logic in one graph, and separating creation from simulation, is very (VERY!) frustrating.
I blame it on my non existing experience and misconception, but I thought it would be easier, knowing what I need to do technically…

I will give it some more try, before I start spamming this thread with questions.

sekow
well for the moment i am struggling a lot.
not having the whole math/logic in one graph,

It can be useful to think of SI and Houdini as musical instruments - one would not play a piano like a cello

hey sebastian

this looks like a great project to wrap your head around houdini… i hope you don't mind if i participate some? ie. attempt to replicate similar tasks?

of course we can share our findings here and help each other get a handle on this new process.

thoughts?

steven

Beautiful work Sebastian.

I would love to see how to approach this the Houdini-way.

Hey Steven, glad to have you around too.
And yeah of course. i would like to invite everybody to participate.

Here is a small prototype so far. clustering went pretty easy, on the nth attempt I must say. The “use high-level nodes” way.
I am going to prepare a full vop sop version tomorrow.

I am so used to work with low level nodes in ICE, that its hard to get a feeling for the way Houdini is presenting itself.
But I am willing to change my purist mind.
Please don't get me wrong, I know this is the right Software for me.

hey sekow (may I call you Sebastian too?),

check this file that will help you with that little doubt regardin the “SRCID” local variable not working.

I offer two ways of fixing it for you to check. I colored the added nodes and note in red.

Actually it wasn´t your fault, but a problem with the Point Generate tool, and it´s very simple to fix.

By the way, a funny thing…after years with Houdini, it´s the first time I´ve ever seen the Point generate SOP, go figure. Must be pretty recent. It looks quite useful though. I tend to use scatter or Point replicate more. Still, because it´s just a small snippet of the project I´m not totally sure of what you´re trying to achieve here.

PS: by the way, in spite of your comments, it looks like you´re doing pretty well around Houdini

I was convinced that I was doing something faulty or misunderstood something. Never occurred to me as a bug …
Thanks Javier for the clarification.

Unfortunately I have to work on a job right now and have just limited time (render/ simbreaks) to figure this whole voodoo out. So this is how far I could go today.
I hope this whole point clustering thing will make sense in the next days, after the weekend.

So, please keep an eye on it. Every help is much appreciated.

-Sebastian

I wouldn´t qualify as a bug a tool that creates an attribute but doesn´t assign it a local variable. More like a missing feature.

I will try to follow this thread.

if you don't mind here are my 2 cents

owlYzarc
if you don't mind here are my 2 cents

I wish i had not a look into it. I wanted to get there myself
Nevertheless, great inside.
Thanks.

-sebastian

oh, sorry then will try to not overdo thing and warn about spoilers in scene files

owlYzarc
oh, sorry then will try to not overdo thing and warn about spoilers in scene files

i don't wanna hijack this but the beauty about using houdini is that there are a gazillion way of doing the same things so you should keep going with spoiler alert.
I have been using houdini for 3 years straight and every day is learning experience. :wink:

If you're interested in a pure POPs example of this type of particle simulation, I've created a .hip file with a few sticky notes to help explain what is happening. This uses the NewPops setup developed for H13.

I attempted to replicate what I thought you were describing ( Clusters breaking apart and leaving streams ).

As was mentioned earlier, there are many different approaches to solving these types of problems in Houdini.

Let me know if you have any questions!

I've replaced your pop wrangle (@activate=fit01(rand(i@id),0,5) with pop vop doing the same thing and I notice in performance monitor that popvop is 2x faster (tested with 6million particles on first replicator on 24 frames)

I assume times will add up with more ‘wrangles’, so Its better to replace all VEXpressions with popvops for performance reasons?

Not really. It depends on how you use the VEX Wrangle.

This issue is very well explained at this tutorial

http://www.sidefx.com/index.php?option=com_content&task=view&id=2512&Itemid=132 [sidefx.com]

where Ari tells how to use Wrangle, and how to get the maximum speed out of your them.

This being said, usually both, VOPS and VEX Wrangles are pretty fast.

What he said!