Hi,
I see a description of GU_Fur in the H11.1 HDK documentation
http://www.sidefx.com/docs/hdk11.1/ [sidefx.com]
Is there a way to retrieve an instance for it for a given Fur SOP ?
I have applied dynamics to the Fur SOP and was wondering if I can use this to extract the dynamic guides and other information.
Regards
HDK : GU_Fur
3589 5 0- nicholas_yue
- Member
- 301 posts
- Joined: July 2005
- Offline
- derrick
- Staff
- 329 posts
- Joined: July 2005
- Offline
The Fur SOP is a thin wrapper around GU_Fur. It basically passes the input geometries and parameters directly to GU_Fur and implements hooks to capture any errors and warning messages produced. The GU_Fur instance used by the Fur SOP is a temporary object used when cooking and does not persist after cookMySop() finishes.
- nicholas_yue
- Member
- 301 posts
- Joined: July 2005
- Offline
derrick
The Fur SOP is a thin wrapper around GU_Fur. It basically passes the input geometries and parameters directly to GU_Fur and implements hooks to capture any errors and warning messages produced. The GU_Fur instance used by the Fur SOP is a temporary object used when cooking and does not persist after cookMySop() finishes.
Observation : I noticed that the number input guides in a Fur SOP needs to match the vertex/CV count of the input geometry, I believe there is some interpolation involved which leads me to believe that those guides are pertubed/sim'ed-with-dynamics before the interpolation.
So….if I am interested in the dynamically affected guides which goes into the interpolation code…..how do I get hold of those guides (presumably sample-able at fractional time for motion blur time sampling)
Regards
Nicholas Yue
- derrick
- Staff
- 329 posts
- Joined: July 2005
- Offline
The Fur SOP interpolates the guide hairs but does not perform any kind of dynamics simulation. Any dynamics would have to be applied to the guide hairs before they are given to the Fur SOP.
The Fur SOP supports two ways of specifying guide hairs. One method requires the number of guide hairs to match the number of points in the skin geometry. The other method (the default when using the Fur Object) lets you place guide hairs at arbitrary positions on a character. The Fur SOP will use the second method if the guide hairs have a float3 primitive attribute called “guideorigin” indicating where the hair is attached on the character in its rest pose.
The Fur SOP supports two ways of specifying guide hairs. One method requires the number of guide hairs to match the number of points in the skin geometry. The other method (the default when using the Fur Object) lets you place guide hairs at arbitrary positions on a character. The Fur SOP will use the second method if the guide hairs have a float3 primitive attribute called “guideorigin” indicating where the hair is attached on the character in its rest pose.
- symek
- Member
- 1390 posts
- Joined: July 2005
- Offline
derrick
The Fur SOP will use the second method if the guide hairs have a float3 primitive attribute called “guideorigin” indicating where the hair is attached on the character in its rest pose.
Wow, thanks for that info! I had a problem with finding the new guides method (scattered) in GU_Fur.
If I may hijack a thread I have a related question. Fur SOP prior H11 allowed to actually scatter guides on curves, which was a nice option for making feathers. It was kind of buggy (making some assumption about curve's area), so I can understand why in H11 it was fixed/disabled. Are there chances to make it happen again, better perhaps? I would love to be able to scatter perpendicular guides along curve's tangent…
thanks,
skk.
- nicholas_yue
- Member
- 301 posts
- Joined: July 2005
- Offline
SYmek
If I may hijack a thread I have a related question. Fur SOP prior H11 allowed to actually scatter guides on curves, which was a nice option for making feathers.
I see some similarity with my quest/query.
If HDK has a unified way to return the per vertex information (guides) regardless of the input method (with/without “guideorigin”) plus a way to query the final scatter location (possibly in UV space of the input surface), it'll allow more interesting work (like feathers, per final strand snow particles) to be done for the the current Fur Object.
My 2cents.
Regards
Nicholas Yue
-
- Quick Links