You can avoid the double lines if you discard not only the equal index distance checks but also where one index is smaller or equal to the other. So if you compare 6 to 3 then it adds a line but 3 to 6 it doesn't. Awesome video! Learned a lot from it :)

@metanoia7217

5 ай бұрын

Another potential idea would be to sample only the even indexed (modulo_2=0) points. So you could cut the iterations by half and by combining your ideas with the one I'm thinking about, you'd never end-up with duplicate geometries while significantly reducing the amount of computations... ;)

@benjoe1993

5 ай бұрын

@@metanoia7217 That would have an effect on the outcome because 3 and 5 wouldn't be connected even if they are close enough.

@metanoia7217

5 ай бұрын

​@@benjoe1993 Thanks for your comment! I forgot about the "odd-odd" connections, sorry :)

@gavindotjs

5 ай бұрын

That's a fantastic solution! Thank you!

Double lines could also be worked out with a nested repeat zones.

fun fact: repeat node = slow not saying that this is a bad approch but pls stop using the distance node and use the length node and ^2 the min distance

@gavindotjs

5 ай бұрын

You're absolutely right that the repeat nodes are slower, and I thought I'd mentioned that in this video, but if not that's why I linked to the other tutorial. I've also mentioned in other videos that the repeat nodes aren't the best way to do this and similar effects, but they can be easier to understand. Also, I'm not sure I follow your math. Distance is required for this effect and is the foundation of it. If I read that correctly, squaring the minimum length between the two vectors you're comparing won't give you the same result as measuring the distance. Sorry if I misunderstood!

@user-mv4cj4gu6b

5 ай бұрын

so you used square root(dx^2+dy^2+dz^2)

✔️ Promo*SM