View Full Version : Ani-reflections alignment

03-28-2007, 06:00 PM
Hi all,
I'm trying to get ani-reflections to work on an object, but the object was modelled in position and its axis doesn't like exactly on x, y or z so placing the axis is a bit awkward.
I've attached an example image of what I'm trying to do that shows a cylinder rotated so its central axis is now quite arbitrary. I've been trying to position the ani-reflections by using to reference nulls at either end of the cylinder, and though I've been able to get the nice looking ani at the top of the object, what I'm really after is the nice 'straight line' style reflections down the sides (pointed out with the blue highlight in the image).

Does anyone have any idea how this is done?

03-28-2007, 06:14 PM
Good question! :)

03-29-2007, 10:04 AM
Seriously, no one knows how to do this?

Dave Jerrard
03-29-2007, 12:32 PM
Have you tried rotating a null to match the rotation of the object, and then copying those rotation values to the Ani-Reflections node's rotation settings? Another option would be to UV the object and use the UV maps to defined the anisotropic directions.

He Who Thinks At Least One Of Those Should Work.

03-31-2007, 04:21 AM
Hi Dave,
I haven't tried just copying the values from the rotation of the null directly, but in the node editor that is exactly what I have done. I took the rotation output and plugged that into the rotation input of the ani-reflections node. This is essentially the same thing, but it produces the effect seen in the pic above. I just dont get why that would be the case.

I dont know how to use the UV option yet as I'm having a hard time finding the documentation that deals with this node..which is daft. How does the UVing work with ani?

04-01-2007, 11:16 AM
I was going to ask about this in the 9.0 beta period, but I forgot...
Anisotropic specularity is the result of a fibrous surface, right? As far as I can see, rotation doesn't rotate the fibres, but the specular highlight itself.

Anyone confirm or cut my theory to pieces? :)

05-31-2007, 09:22 AM
I'm no expert, looking at Dan's inside v9 edition (my latest LW expense!)
page 449, there abit about Anisotrophy U &V coordinates that might be the answer, squeezing the U/V ie 90%,10%. Sorry Dan, I mean buy the book, I'm finding it a great deal of help and I've only had it 5 minutes.