I'm Following a fracture tutorial build in Houdini 12.5 with my Houdini 15 in which scatter node doesn't have Attribute bias to control the range of density based on painted color.

Here is my tree of nodes

I imported an OBJ model and connected it to a point node,in which the color channel is set to add with a value of 0,0,0, then i added a paint node ,and painted some red color with the value of 1,0,0, to make some variable density for the scatter node at the bottom.

However, without having the Attribute Bias in my 15, I couldn't achieve to control the density range based on Cd, as the tutorial had done. I got a result of only the red painted region being scattered with points, and black region with none points. I want the black region also to scatter points, but less than the red parts. and the control the proportion between these two values with a slider. how can I achieve this in Houdini 15?

Sorry for my poor expression, i'm brand new to Houdini and am not an English speaker.

Thank you

Just don't go all the way to black with your color. Instead of black choose grey then move to black. Then you will get some points in that grey region as well. Make sure to set your Density Attribute inside the Scatter node to Cd, it defaults to density.

Enivob
Just don't go all the way to black with your color. Instead of black choose grey then move to black. Then you will get some points in that grey region as well. Make sure to set your Density Attribute inside the Scatter node to Cd, it defaults to density.

Thank you Enivob, but I want to control their proportion more procedually with some slider instead of turning back everytime to the color to get the sweet spot I wanted.

What this bias essentially is, is a blend between your attribute and 1.0.
So you can easily build your own bias with a wrangle or point VOP

You are also not limited to Cd and it's a good practice to use your custom attribute to do what you want.
There are some nodes that override color and it might get you in trouble at some point.
And in the case of the density attribute: you just need a float, so there is no need to store a vector.
You can always preview any attribute with a visualizer or a Color SOP ( ramp from attribute )

blackpixel
What this bias essentially is, is a blend between your attribute and 1.0.
So you can easily build your own bias with a wrangle or point VOP

You are also not limited to Cd and it's a good practice to use your custom attribute to do what you want.
There are some nodes that override color and it might get you in trouble at some point.
And in the case of the density attribute: you just need a float, so there is no need to store a vector.
You can always preview any attribute with a visualizer or a Color SOP ( ramp from attribute )

Thank you very much, Mariusz Wesierski, your example is very helpful for me.

blackpixel
What this bias essentially is, is a blend between your attribute and 1.0.
So you can easily build your own bias with a wrangle or point VOP

You are also not limited to Cd and it's a good practice to use your custom attribute to do what you want.
There are some nodes that override color and it might get you in trouble at some point.
And in the case of the density attribute: you just need a float, so there is no need to store a vector.
You can always preview any attribute with a visualizer or a Color SOP ( ramp from attribute )

Great! Smooth node makes it very fancy!
Thanks!!!