Different result, the same logic. Why?

[bloomendale]

Hi.

I've made 2 scenes with identical logic ( as i thought).

1) The first one - i find the new velocity vector in sopnet in dops (red one) and use it in Motion DOP. While it sometimes colors some node in red (error) i just recook them and get proper result - cubes are forced to go to [0,0,0]. Cubes behave as it should be.

2) In 2nd scene i use sop solver to calculate velocity vector and use it in Motion DOP.

Math behind these 2 setups is very simple - just negate current position.

But cubes starts to behave strange after ~150 frames. And it solves slower than 1st example.

Why this happens?

And what is the proper way to make some calculations for DOPs in SOPs, using some data from DOPs? If expressions are not enough.

For example i need Position data of the DOP object to use in complex calculations of its speed in SOPs (VOP SOP etc) and then bring the result back and apply it in DOPs.

As i see the difference between these 2 scenes is that in 1st i calculate velocity before Solve stage of the DOPs - so i have Calculate velocity (red sopnet) -> Apply it (Motion DOP) -> Solve RBD objects with rbd Solver.

and in 2nd - i have Apply previous velocity (Motion DOP) -> Solve new velocity (calculate in SOP Solver DOP) -> Solve the RBD objects.

Am i right?

Why i can't move my red sopnet in 1st example outside AutoDopNetwork? It fails to solve.

BIG thanks in advance.

3ak_moveToGeoPonts1.hipnc

3ak_moveToGeoPonts2.hipnc

[bloomendale]

Anyone??

btw, i found why in sopsolver's example spheres are flying strange - we need to turn off attribcreate node in sopsolver with "forcet" attrib - the one i use in Point Force DOP. But WHY this is have such impact?

I hope someone answers the main question (in bold in previous post).