I would like to be able to view the velocity created by a trail SOP. I know you can map via point SOP > add velocity > $NX , $NY , $NZ but of course this is normalised.

any ideas ?

r

You can visualize any attribute now (since 9.0). In the viewport display options add a custom attribute display. VERY handy.

I have done this , under custom display $V is already there , check it on , there's nothing to visualise. Add a point SOP and map the normals to the add vel tab and it works and of course I dont want to add $V I want to see what's already there.

r

I'm not sure if you are being literal, but the attribute is called ‘v’ – and that's what you want to plug in under the field. you mentioned ‘$V’ in your posting which will probably not work. (I don't have houdini in front of me at the moment, so I can't check your file…)

G

yeah sorry about that bit of confusion v is what I am talking about. I saw an interesting post where somebody wanted to take their animated geometry and use the v attribute from the geometry to drive the vel in a fluid sim. Hense my interest in seeing just how much velocity my geometry is creating

r

So I finally had a chance to take a look at your file…

What's happening there, is that the velocity is calculated BEFORE the world space xform is being applied. ie: the as far as the trail sop is concerned, your torus is completely static. The sop chain cooks, then the geometry is transformed.

Two ways around this:

(1) apply the transformation at the sop level (it becomes a deformation at this point) – then cook the trail sop.
(2) object_merge your world-space xform'ed into a new object, and cook the xform there. again, a deformation at this point.

Holler if you need a sample file.

G

ps: your OTHER sanity checker in situations like this is the spreadsheet. If you pop it open and look at the ‘v’ attribute in your existing file, you'll see that it's all 0's – which lead me to look elsewhere for why this velocity wasn't being calculated as expected.

Hi ,
Thanks for the reply after much fiddling I cam to the same conclusion and I even stumbled across the object dop ( very handy ).

r

circusmonkey
…. SOP > add velocity > $NX , $NY , $NZ but of course this is normalised.

It shouldn't be normalised at all. What you described is perfectly valid and works.