Opposite attractor

Attractors based on distance naturally tend to make the the thing they are closest to smaller because the distance is smaller. Anyone know how to reverse this so the closer items are bigger? Is it a situation of reordering the list?

Replies to This Discussion

Permalink Reply by David Reeves on August 24, 2011 at 12:51am

Hi Michael,

1/d would give you an inverse relationship. As d approaches 0 you'll get a number that approaches infinity. If you do expect to have distance values of 0 you'll want to add a constant to d before inverting it to avoid a divide-by-zero error.

Permalink Reply by Peter Eliason on November 2, 2011 at 6:41pm

I'm running into the same issue- were you able to figure it out so that the object become larger near the points?

Permalink Reply by Michael Pryor on November 2, 2011 at 7:24pm

use the 1/x component

Permalink Reply by Peter Eliason on November 2, 2011 at 8:23pm

This is the definition I'm working with, but it doesn't seem to be coming up...

Attachments:

DEFINITION 11-1.jpg, 261 KB

Permalink Reply by Michael Pryor on November 2, 2011 at 9:07pm

use remap as Mateusz suggested

Permalink Reply by Mateusz Zwierzycki on November 2, 2011 at 7:49pm

remap values -> source domain i.e 0 to 1 , target domain 1 to 0

Permalink Reply by Michael Pryor on November 2, 2011 at 8:22pm

I always forget about that.

Permalink Reply by Sophia Keivanlo on March 5, 2012 at 11:12am

Sorry, I have the same problem, and I cant figure it out, my circles all become very small and the same size.. Where do you put the remap? Do you use it together with 1/x or on it's own?

I tried many different things, this is how the mess looks like:

Could you please post a definition so that I can learn it?

Over and out!/Sophia :)

Permalink Reply by Sophia Keivanlo on March 5, 2012 at 11:24am

Ok, I cleaned it up a bit to clarify what I mean..

Before 1/x

and after .. (??!?!?!!?!)

why do they all become so small? I tried to multiply the radius after, but it didn't work..

Any hints?

/Sophia

Permalink Reply by Hannes Löschke on March 5, 2012 at 12:11pm

This is most likely a scale issue. When you get distances like 500 units and multiply by 0.1 to get useful radii, you end up with 0.002 after the 1/x where you had 50 before.

Permalink Reply by Michael Pryor on March 5, 2012 at 12:14pm

yea, it also isnt true to the proportion of scaling, I just stick now with remap.

Permalink Reply by Sophia Keivanlo on March 5, 2012 at 12:42pm

Thank you so much for taking your time to answer this!

I don't know why I can't make it work.. Have I missed something obvious?

The green points are the ones who should have the biggest circles around them, and the others should fade out gradually..

Source domain is set on 0-1 and the other on 1-0