Kangaroo vs. Kangaroo2

kangaroo
unhandled

#1

Hi,k1VSk2_v1.gh (176.5 KB)

I’m wondering if anyone might explain the difference between the forces/goals in K1 and K2?
Please find attached an example with the two producing seemingly very different geometries from similar nodes.

Thanks!


(IVELIN PEYCHEV) #2


#3

Hi Ivelin, thanks. I’ve read through that before but it doesn’t really answer my question.

The taller geometry was obtained from K2 while the shorter K1:


(IVELIN PEYCHEV) #4

Hi @timothytai,

Yeah, sorry, that’s as much info as I can give you. I just saw you had no answers and tried to help as much as I can.

I tend to stay away from Kangaroo (either versions).

Good Luck,
Ivelin


#5

Both results seem to be pretty much the same to me.(K2=Much faster)

k1VSk2_v1_re.gh (40.6 KB)


(Daniel Piker) #6

Hi Timothy,

The difference between your simulations was the rest length of the edges. Once these are made the same, both solvers converge to exactly the same geometry.

The mouse-over description of the Length input in the K2 ‘Length’ goal - "Length - if none provided, starting length will be used"

Whereas for the “RestLengthFactor” in the SpringsFromMesh component in K1, you had 0, meaning all edges had a rest length of 0.


#7

@HS_Kim and @DanielPiker thanks a lot!
Overlooked the “RestLengthFactor”; now realized if I want the same result from K1, I’ll need to input 1 instead of 0.