Volume - Sample attr from point - Strange behaviour

[BenWall]

Hey there,

After spending a fair amount of time on SOP and VEX, I'm now learning basic volumes handling. I found a behavior I don't really understand and can't find any ressources on the subject.

I created a simple setup to change the density of a scalar volume with the distance to a point. Everything works as expected, but when I move the point to a bound of the volume, the density from on side seems to be "reflected" on the other side of this volume. Looking at the screenshot, we should only get yellow values on the left (where the point is), the right side should be blue.

Seems to me there are three possibilities here:

- My volume wrangle might be wrong,

- There is something I didn't understand about volumes,

- There is something wrong with the way I setuped the volume visualization.

 

Can someone explain what I didn't get right?

Thanks

strange_density.hip

[Librarian]

Fit Function.. You have Reverse Or?

[BenWall]

Not sure I understand your question.

I used fit() in order to use chf("max_dist") and invert the values at the same time. Is there a better way to do this?

[vicvvsh]

Hi,

i think it's just volume visualization node's viewport visualisation imprefection.

[BenWall]

Narrowing it down to avoid vex errors with a distance check, I still get the error
 

float dist = distance(@P, point(1, 'P', 0));
float d;

if(dist < chf('max_dist')){
    d = fit(dist, 0, chf('max_dist'), 1, 0);
    d = clamp(d, 0, 1);
    d = chramp('gamma', d);
    f@density = d;
}else{
    f@density = 0;
}

 

[Librarian]

@BenWall Hm near-point-Function Maybe .
 

strange_densityDnerpoint.hiplc

Edited January 16, 2021 by Librarian

[BenWall]

Seems to work like a charm with nearpoint(), thanks.

Double checking by trying to get sample density on a point near the side which is not supposed to have any density, the data seem correct. I think vicvvsh is right: it's a volume visualizer error.

[Librarian]

Seams like Error Exact