Thanks Guys, Turns out the reason the envelope operator was being ignored was something to do with the execution order on the object. For some reason the enveloped object had at some point had a simulated tree on it so had all the simulation stack entries. As soon as these were removed (new object) problem went away.
Cheers, Jules On Tue, Mar 12, 2013 at 1:15 AM, Edy Susanto Lim <[email protected]>wrote: > Hi, > Setting the envelope weight using the per deformer weight seems to work > just fine as long as you know the deformer index. > In the attached scene, I use a null (E_ref) to define a new weight area of > the new influence null (E). > > Cheers, > edy > > > > On Tue, Mar 12, 2013 at 4:54 AM, Jules Stevenson <[email protected] > > wrote: > >> Yep, just worked this out :). The whole process of envelope generation is >> automatic, so ordering shouldn't be a problem... I've still got issues with >> the rig not actually deforming the geo, even with the DQ / ice based setup, >> but working through it. >> >> >> On Mon, Mar 11, 2013 at 5:40 PM, Alan Fregtman >> <[email protected]>wrote: >> >>> Yeah, and if you delete the Envelope Op it gets confused when painting. >>> It's rather lame that you have to mute it. >>> >>> I should've clarified I didn't mean to delete the op itself, but mute it >>> and use an ICE equivalent instead. >>> >>> >>> >>> >>> On Mon, Mar 11, 2013 at 1:33 PM, Peter Agg <[email protected]>wrote: >>> >>>> Even with the DQ compound you still have to paint a regular envelope, >>>> mute the operator and replace it with an ICE one. You also have to be very >>>> careful about the group ordering you give the compound as, if the order >>>> changes, it can get a bit funky! >>>> >>>> Basically I wouldn't really recommend it unless you need DQ. :) Still >>>> fun to check out. >>>> >>>> >>>> >>>> >>>> On 11 March 2013 17:23, Jules Stevenson <[email protected]>wrote: >>>> >>>>> Ahh, interesting, thanks Alan. Will have a look... >>>>> >>>>> >>>>> On Mon, Mar 11, 2013 at 5:11 PM, Alan Fregtman < >>>>> [email protected]> wrote: >>>>> >>>>>> Perhaps ditch the classic EnvelopeOp for the ICE envelope equivalent? >>>>>> The "Dual Quaternion Deformation" compound can do Linear (classic) Blend >>>>>> Skinning. >>>>>> >>>>>> If you want to ditch DQ altogether you can dig into the compound and >>>>>> strip out the DQ logic, leaving only the linear bind. >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> On Mon, Mar 11, 2013 at 5:58 AM, Jules Stevenson < >>>>>> [email protected]> wrote: >>>>>> >>>>>>> Hey Gang, is there any documentation for correctly setting envelope >>>>>>> weights via ice? >>>>>>> >>>>>>> I can see from dialing down the available properties of the cluster >>>>>>> you have the per point envelope weights per deformer array, plus another >>>>>>> attribute of per point envelope weights (so not referenced against the >>>>>>> deformer index array, I'm presuming this is just for data >>>>>>> visualization). >>>>>>> However when I set these I can see the weight colours pop into >>>>>>> existence in >>>>>>> the view-port, but the actual envelope operator does nothing, it fails >>>>>>> to >>>>>>> bind these new weights to the object - the mesh stays still as the >>>>>>> deformers move. >>>>>>> >>>>>>> Is there a best practice involved here? >>>>>>> >>>>>>> Many thanks for any help, >>>>>>> >>>>>>> Jules >>>>>>> >>>>>> >>>>>> >>>>> >>>> >>> >> > > > -- > Edy Susanto Lim > TD > http://sawamura.neorack.com
