When a groom is allowed to simulate, only the strand representation computes correct world normals for the new hair strand positions. This is especially noticeable when the groom is inverted as gravity causes substantial non-skinned movement.
Skinned movement of the groom appears to be transferred to the cards correctly.
The easiest way to reproduce this issue is to use a long hair groom with simulation enabled and turn it upside down. If the hairs are not falling sufficiently far to make the issue readily apparent, reducing bend stiffness can help.
In the world normal buffer visualization, the hair strands will update their normals during simulation. When the camera is moved away, the LOD of the groom will change to cards, which will not have their normals update and will differ significantly from the strands.
Expected: general orientation of the normals to be similar between strands and cards
Actual: card normals appear to not respect the new hair positions caused by the hair falling under simulation.
N/A
There's no existing public thread on this issue, so head over to Questions & Answers just mention UE-247749 in the post.
| 0 |
| Component | UE - Graphics Features |
|---|---|
| Affects Versions | 5.5 |
| Target Fix | 5.7 |
| Fix Commit | 40785633 |
|---|
| Created | Feb 13, 2025 |
|---|---|
| Resolved | Mar 17, 2025 |
| Updated | Apr 17, 2025 |