I think it would be better to use a color at vertices map instead. Have a different color represent a different feather type. Weight maps are going to be harder since your going to be most likely dealing with 0-1 scalar values which will need to be rounded to pick each instance from the group. Since you have 15-20 instance members your value differences on your weight map are going to be minute and very tedious to work with. At least with colors it can be more visually obvious there is a difference between two vertices.
Could also paint a texture map with in the same line of thinking and look that up from the feather emit location to pick the instance based on the painted color too.
On Dec 28, 2011 5:34 PM, "Chris Covelli" <kylev...@gmail.com> wrote:
Sorry to bring this up again,
We've decided instead of converting the particle feathers to geo, we're going to texture the instanced geo feathers individually, which has increased the number of instanced feathers significantly.
My new question is; If we are using a weight map to drive the "Type" parameter, is there any way to tell what weight values correspond to which instance group member? Its easy enough to feel it out if you're only using 3 or 4 different kinds of feathers, but when you've got 15-20, It be nice to have some sort of point of reference.
Thanks again!!
-Chris
On Tue, Dec 20, 2011 at 10:22 AM, Chris Covelli <kylev...@gmail.com> wrote:
>
> Hi Sandy,
>
> Th...
another approach that I recommend is to use a dummy envelope to get the
index, and you can easily paint the influence with this method.
see the attached image.
Scene file : http://sawamura.neorack.com/temp/deformer2instance.rar
Cheers,
edy
On 12/29/2011 11:42 PM, Chris Covelli wrote:
> Cool! Ill give the color at vertices map a try.
>
> Thanks!!
>
> On Thu, Dec 29, 2011 at 1:11 AM, Eric Thivierge <ethiv...@gmail.com
> <mailto:ethiv...@gmail.com>> wrote:
>
> I think it would be better to use a color at vertices map instead.
> Have a different color represent a different feather type. Weight
> maps are going to be harder since your going to be most likely
> dealing with 0-1 scalar values which will need to be rounded to pick
> each instance from the group. Since you have 15-20 instance members
> your value differences on your weight map are going to be minute and
> very tedious to work with. At least with colors it can be more
> visually obvious there is a difference between two vertices.
>
> Could also paint a texture map with in the same line of thinking and
> look that up from the feather emit location to pick the instance
> based on the painted color too.
>
>> On Dec 28, 2011 5:34 PM, "Chris Covelli" <kylev...@gmail.com
You and your envelopes... :)
On Jan 1, 2012 9:53 PM, "Edy Susanto" <edysus...@gmail.com> wrote:
not sure about that. but if you are using colors on vertices, there are a couple of obstacles:
- filtering color to decide which feather would mean that you have to compare 4 values just to get 1 integer value.
- as color on vertices works per vertices, and if your feather is grown on polygon surface and not per point, you'll have issues with the blended color. one way to tackle this one is to get closest vertex and read the color.
another approach that I recommend is to use a dummy envelope to get the index, and you can easily paint the influence with this method.
see the attached image.
Scene file : http://sawamura.neorack.com/temp/deformer2instance.rar
Cheers,
edy
On 12/29/2011 11:42 PM, Chris Covelli wrote:
>
> Cool! Ill give the color at vertices map a try.
>
> Thanks!!
>
> On Thu, Dec 29, 2011 at 1:11 A...
> <mailto:ethiv...@gmail.com>> wrote:
>
> I think it would be better to use a color at vertices...
>> On Dec 28, 2011 5:34 PM, "Chris Covelli" <kylev...@gmail.com
>> <mailto:kylev...@gmail.com>> wrote:
>>
>> Sorry to bring this up again,
>>
>> We've de...
>> <kylev...@gmail.com <mailto:kylev...@gmail.com>> wrote:
>> >
>> > Hi Sandy,
>> >
...