A bit late with this answer, but octanerender mainly uses wavefront .obj files as input, and their materials is incompatible with anything at all(that's expected tbh), so basically all you need is to export .obj and tweak materials inside of octane.

They have free demo on their site, so anyone can test it

Aha, i wasn't sure if houdini embeds OTLs into hip files or not. Here's hda.

Basically it's a very simple procedural lightning with randomized seed based on destination point coord, so every lightning in a pack differs from others.
Probably nothing new for you guys, it was made to refresh my houdini knowledge. Probably i will make more advanced version with forking as a python sop soon, seems it's easier for to just write little lightning(src, dst) function than make all these node trees.

Here's the results, i was doing that just for training purposes, think the next one is gonna be easier to implement as a python sop.

Exactly. By the way i didn't know “add” can also make lines.

Thanks for help!

Actually….i just thought that i probably could stamp point coordinates in that asset, so seems copy is totally usable. Going to try it as well.
Thanks for input guys, i'm gonna post that thing i'm doing when everything will work

It's not a simple copy, it's a dynamic asset that uses 2 input points, so i have to know position of each point im iterating in space.

in Maya i would do something like:

for($i=0; $i<$numPts; $i++)
{
float $a = `xform -q -a -ws -t pSphereShape1.vtx`;
line(0,0,0, $a, $a, $a);
}

Examining your files at the moment, just want to explain: imagine a simple grid wired into foreach sop, what i want, is, say, to create sphere or box at each grid point location.
Just for some reason i cannot wire network to “each” inside, and have to get that point coords with expression.

I'm digging information about foreach and seems i'm starting to understand how it works, but 2 things still unclear:
1. how to get object connected to foreach inputs? opinput() or something else?
And is it possible to refer to those input “sockets” from inside the foreach?
2. how to get coordinates of a point currently processed inside of foreach sop?
I've got a number of groups with single point in each, made with “partition”, im iterating over these groups, but how to get position of that single point that is currently being processed?

Update: fixed mistake, partition, not attribcreate. Attribcreate i tried for iterating over point attributes.

By the way: is it possible to access “composite parameters” in houdini, like in maya where you can do “setAttr box1.tx 0.1” or as well “setAttr box1.t 0.1 0.2 0.3”?

At the moment i have quite a long multiline python expression, and i have to repeat it with minimal differences for tx/ty/tz.

Bingo! It works! Thanks a lot, free beer from me

circusmonkey
I am talking about the display on the normal on points

I'm talking about the same: turning on “v” in display options either crashing H12 or shows all the impact velocities pointing in the same weird direction.

Guess that's why i'm getting weird velocity vectors from impactAnalysis node.

Here is it. Basically i'm trying to create a line between 2 points, but since a “line” sop doesn't have any inputs, i have to use expressions to access geo translate channels. The reason i'm doing it that way is that i want to make a preset/hda to reuse it later by just changing sop paths.

Guess that's gonna work, but is there some more elegant and simple solution? I mean i thought that per-point emission rates is quite a basic thing, why so many problems with it…

I've ran into problem with expressions and after hour of experiments im giving up: i created a node and added 4 parameters to it: 2x op paths, string and float.
What i want is to enter path in “sop path” param, get full path with channel name in “test string” param and get value of that channel in “test float” param. Op paths works as expected, after some time i figured out how to build full path in “string param: chs(”p2“)+”/tz“ + key on that channel.

Now - how to get the actual value of that channel?
Tried ch(ch”tstr")), but it doesn't work.
Any suggestions?

Let me explain the effect i'm trying to achieve:
It's a simple scene with RBD box falling onto RBD ground. Box has impactAnalysis attached, so i have the impact points in “import_impacts” geometry network.
What i'm actually trying to do is to emit particles from those points with popnet, with “source” node. It's actually work, but at the moment i'm getting constant stream of particles from each impact point, so guess what i need to do is to get impacttime of each impactpoint, and filter them with simple condition: if($T-$impacttime < 2) to emit particles just from those points that is 2s old or less. And i think i can scale actual emission amount by “impulse” attribute. That's why i need per-point emission rates.

Besides, there's also a vop sop that was supposed to reflect impact vector to generate new particles in proper direction, but atm i cannot even find that direction, since it seems that “v” holds some arbitrary data, and i cannot find any RBD velocity attributes.

Spent whole evening tryint to visualize vectors with “point/add” and polywire nodes(since custom display one is simply crashes H12), but encountered so many problems with expressions so i'm almost thrown my laptop out of the window.

Doesn't putting $NPT/using npoints() will emit “number-of-points” particles every second?
What i wanted to do is take some pre-created attribute on source geometry, and emit particles according to it. Say, “point N7 has attribute ”emit“ equal 13 = we need to emit 13 particles from this point”

I have an arbitrary point cloud with some custom numeric attribute that was supposed to be used as a per-point emission rate value.
The thing i cannot find is actually _how_ to emit required amount of particles from each point per time step/frame? Any ideas?

I think they developing that stuff for PhysX library, probably going to be introduced at the next Siggraph.

Quick question about impactAnalysis: does v vector in impactAnalysis data represents actual “direction of impact”? I've tried to visualise that vector in viewport, but it seems all vectors on all impactPoints always look into the same direction, and it's obviously not the direction of impact.

If not - do you guys know how to get the direction of impact?