Thanks, yea, I tried a tube after I realized line was actually just an infinitely thin line. no luck so far with the tube even after making it really wide. gonna try a grid. that might work better.

I'm rendering a scene with pbr and have an area light of type line (to represent long fluorescent light fixtures) that has raytraced shadows. I have given it an area size of 20 x 50 and have attenuation turned to physically based. But I still get very hard shadows.

Is there something I can do to soften the raytraced shadows?

thanks for the tip! Not ideal, but it is working!

edward

tettoffensive
When I'm running a simulation by just hitting play in the timeline, I would like to set breakpoints (just as you would in a debugging software) rather than constantly changing the last frame. So say my sim goes from 1-100, I might want to set a breakpoint at frame 5 and 20. So it would stop at frame 5 so I could check it to make sure that's what I want and then hit play to continue.

A hacky way to do this is to create a dummy Null node and create a single channel on it and pin it in the Channel Editor pane. Create keys at frame 5 and 20. Now you can use the “previous key” and “next key” buttons on the timeline to go to them.

I haven't found any documentation and not much help with the smoke_mask and fire_mask. I did see that they might be broken in some versions of 12.5? But I'm finding that the smoke_mask is mostly white and fire_mask doesn't show much even though I have areas of very hot/bright fire.

I do have the smoke density cranked up so I get very dense smoke but I have also modified the shader to take the density and multiply it by the inverse ramp of heat so that it doesn't cover the fire.

But I'm finding that the more I multiply down the density in these areas which reveals more fire in the beauty pass, the smoke_mask get's whiter instead of blacker.

Anyone know how this vex code works that controls the smoke_mask? Or maybe I'm just missing some settings on the image plane of the mantra node?

Thanks

If there's already a way to do this I haven't found it but would like to know.

When I'm running a simulation by just hitting play in the timeline, I would like to set breakpoints (just as you would in a debugging software) rather than constantly changing the last frame. So say my sim goes from 1-100, I might want to set a breakpoint at frame 5 and 20. So it would stop at frame 5 so I could check it to make sure that's what I want and then hit play to continue.

nanocell
We've been experiencing this rather frustruting problem too, and it seems to stem from the “Delay Inactive Viewport Redraws” setting.

After disabling that setting (Edit -> Perferences -> 3D Viewports -> Delay Inactive Viewport Redraws) the problem seemed to have gone away on this side.

CheerZ,
Van Aarde.

I was wondering if it was this setting. I also noticed that toggling the Scene Renderer to H11 and back to OpenGL 3.2 got rid of it.

So I have a smoke object in a pyro solver and there's a collision object that moves through the smoke. Right now it drags the smoke along with it which is nice, but I want the smoke to “roll” tangental to the direction of the objects velocity and diffuse a little more than just go in the same direction.

Is there a good way to achieve this?

So I have a vop sop that is running a grid of points through an oscillation vop that is moving there position along the normal up and down. I want the speed of this to slow down over time and have keyframe animated a multiplier on the time in seconds. However, as it's animating there is a lot of popping as it slows. I guess because it is jumping to different points on the sine wave. I need it to slow down relatively quickly. Like there's a huge drag force on it. But want to avoid using a popsim.

Any suggestions?

lisa
What magic incantation does KDE need to make houdini -span work properly? Currently it just places the houdini window on the 2nd monitor at startup.

Gnome desktop does the same thing. Any solution for this?

I have a chop network where I take some static geometry and modify it's tz channel with a wave chop. I've done this as I've seen on several examples here via a lookup chop.

The thing that I want to add to this is to have the wave “ripples” start when a trigger goes off. I have a trigger sop that goes off when an attribute reaches a certain value. But I don't know how to make that trigger drive the animation (amplitude/phase) of the wave.

Thanks.

Rafaels-this should be useful. though i think the math is more complicated since the metaweight depends on how far the point is from the center of each metaball. This may be tricky to figure out.

Claudio-this looks useful too. But I have a ton of metaballs so I'll have to play with it to see how heavy it is.

Thanks.

Is there a way I can find out when two or more metaballs intersect? Via an attribute possibly? I'm trying to do a little procedural rippling (like a little shockwave) when 2 or more metaballs combine. I've got a bunch of separate metaballs (done with a copy sop onto a bunch of points w/ pscale). I scale them all up until it is one big piece of geometry.

Thanks.

This will help for sure! Thanks. Now I just have to figure out how to get this to work with N amount of volumes (since I have a volume primitive for each of the voronoi pieces).

I have a bunch of pieces of geometry (from a voronoi fracture) and I want each one of the pieces to turn into a metaball representation. My first attempt was to scatter points and use a copy sop, but that doesn't really give me accurate representation of the pieces without having a lot of points (making it super slow!).

Since the isooffset sop has a metaball mode, I was wondering if there's a way to get the metaball representation as output. I need this so that I can merge the pieces together at some point.

Is there a way to do this?