Timeline for Compute on all points in geometry nodes
Current License: CC BY-SA 4.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 27 at 23:15 | comment | added | Metal3d | As I'm working with a points cloud, I need to iterate over the points IDs, so at this time, a repeat zone is (I guess) mandatory. I'm currently testing Accumulate Field node to understand what is the usage. | |
| Feb 27 at 18:28 | comment | added | StefLAncien | The second solution based on a Cartesian grid is most likely redundant with Markus von Broady's proposal from a mathematical point of view. But as the grid could be permanent outside the simulation zone, the creation/deletion time of the point cloud could be saved. Unless Blender is caching in memory every iteration... | |
| Feb 27 at 18:19 | history | edited | StefLAncien | CC BY-SA 4.0 |
Add another GN graph.
|
| Feb 27 at 13:36 | history | edited | StefLAncien | CC BY-SA 4.0 |
Add links to documentation.
|
| Feb 27 at 13:25 | comment | added | Metal3d | I don't know how this node works ;), I will check the documentation. Just one note: you don't need to sample index for the current point, the "position" attribute is the current one. At least, on my setup, it works. Thanks again for your advices, I'll check the attribute statistic. | |
| Feb 27 at 13:20 | comment | added | StefLAncien | The Attribute Statistic node should be optimized compared to a Repeat Zone. An Accumulate Field could also be tested as it is dedicated to summation; but it is producing extra data not required in this case. | |
| Feb 27 at 13:14 | vote | accept | Metal3d | ||
| Feb 27 at 13:14 | comment | added | Metal3d | Damned, I just found approximately the same solution using a repeat zone inside the simulation getting the domain count and sampling index. Thanks a lot for your answer, I'll take a look on your method to see if it's not more optimal. | |
| Feb 27 at 13:02 | history | answered | StefLAncien | CC BY-SA 4.0 |