Timeline for (Geometry Nodes) Sample curve tangent with multiple splines
Current License: CC BY-SA 4.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 25, 2022 at 14:01 | comment | added | Gordon Brinkmann | Let us continue this discussion in chat. | |
| May 25, 2022 at 12:35 | comment | added | Robin Betts | @Gordon A fully-defined orientation requires 2 vector-determined Euler rotations. One to align an axis, (here, Z) and once that axis is good, another to spin around it. This pair adopts the definition of XYZ used by to-points 'Rotation', and is, on the whole, near identical.. there just seems to be a difference in interpolation at the end-points. i.e I was just trying to figure out what a point's 'rotation' is, if it isn't alignment to tangent and normal. It's no great shakes. | |
| May 25, 2022 at 12:00 | comment | added | Gordon Brinkmann | @RobinBetts First of all, why exactly are you using 2 Align Euler to Vector nodes, one with the Tangent and one with Normal? And the second with Z as pivot? The latter tries to align the X of the object to the normal with restricting it to the local Z axis, which messes up the alignment. The other thing is, aligning twice to different vectors and axes cannot give the same result as using the rotation of the points, since they only have one rotation, which is (in the curve's direction) forward is +Z, to the right is +X and downwards is +Y, whereas the curve's tangent is aligned different. | |
| May 25, 2022 at 11:42 | comment | added | Robin Betts | @Gordon ..zackerly. I would have thought the results should be the same? But the endpoints and penultimate points are rotated differently with the 2 methods.. some difference in interpolation... | |
| May 25, 2022 at 11:36 | comment | added | Gordon Brinkmann | @RobinBetts Not sure if I understand what you mean with different methods of alignment... Align Euler to Vector is calculating the Euler rotation to align to a certain vector. Curve vertices are not rotated (apart from a tilt maybe). After converting with Curve to Points, the generated points are already rotated aligned to the curve, which is why you can use their Rotation directly to align instances. The original curve's tangent and normal get lost, but the Curve to Points node generates these vectors based on the points. These could also be used with Align Euler to Vector. | |
| May 25, 2022 at 9:39 | comment | added | Robin Betts | Phrased much better than I was about to .. :) I'm still trying to reverse-engineer this difference in behaviour between 2 methods of alignment | |
| May 25, 2022 at 7:32 | comment | added | Gordon Brinkmann | @RicoHolmes You're welcome, at first glance I didn't know what was wrong either until I remembered what the Sample Curve node does... | |
| May 25, 2022 at 7:03 | vote | accept | Rico Holmes | ||
| May 25, 2022 at 7:03 | comment | added | Rico Holmes | Oh my goodness, THAT is the answer. Apparently I don't have enough reputation here to upvote your answer, but you get massive appreciation. Thank you so very much. :) | |
| May 25, 2022 at 6:56 | history | answered | Gordon Brinkmann | CC BY-SA 4.0 |