Hi, guys. There are already several topics on the issue, but none of them seems to find working solution. It’s just amazing that the built in node was not fixed in so many dynamo versions already. The most common workaround is scaling the geometry. However, that’s only able to fix a small part of the missing segments problem, as you can see on the picture:
Even though I scaled the geometry 1000 times (while the original surface was already quite big - around 500 x 500 meters), there are still missing segments. I tried adding random zeroes to the scale factor, also without success.
It seems like the scaling method resolves only small segments in the interior, while curves, connecting some of the dots to the outside boundary remain missing.
Anyone has any idea?
I think you are scaling at the wrong point.
Your UVs shouldn’t be limited to inside the domain - that is use values between say -0.1 and 1.1 (or more, depending on how big you want things to go).
Hi, Jacob. I’m not sure I understand. Could you please give an example? In my case, I am trying to generate Voronoi, based on particular points, not just random uvs.
Go back to where you created the UV parameters for the voronoi. They should be numbers, and my guess is they range between 0 and 1. Try expanding those values between say -0.1 and 1.1 and see what happens.
Ok, I get it. Let me give it a try
Thx! You helped once again. Now I see where the issue was:
Perhaps by stretching the uvs further from the surface, I will also get rid of those empty corners. You were right. all my coordinates were in the range 0 - 1. Cheers!
Stretching will help remove those corners, but eventually you create more. Assuming there is some degree of randomization in your graph, they are somewhat inevitable. Instead it’s likely best to accept them as is, or filtering them out.
1 Like
Yeap, I stretched to -0.5 and 1.5 and they seem to be almost gone in most cases. Later, I will rearrange surfaces to match points list, so empty corners will be removed anyway.
1 Like
