- 2019-12-12 at 11:32 pm #21922
Error Message: “v3d.BufferGeometry.computeBoundingSphere(): Computed radius is NaN. The “position” attribute is likely to have NaN values. ”
I have an animation with 36 Shape Keys.
It plays, but does a very strange glitch, inverted shape at the beginning (and that’s where i get the messages).
some of the position values were “-0” in the Y so i fixed those. still get the error.
i found that it kept listing “ShapeKey36” over and over in the attributes of the console details so i deleted that shape key and channel…
but then the errors started saying “Shapekey35″… soo…
what’s the best way to deal with this type?
verge 2.17 pre2
firefox latest2019-12-13 at 7:20 am #219272019-12-14 at 1:21 am #21944
I still get it in Sneak Peak.2019-12-16 at 7:43 am #219852019-12-30 at 11:39 am #22461
got a link to the project?
i fixed it.
i had some seriously messed up F curves.. had to set my default keyframe interpolation to be linear / vector.
but let me tell you..
you have the limited number of shape keys per object as 6 in the manual.
i have over a dozen objects having over 36 shape keys each working fine in my Verge3D project.
…is the number limit a barrier to having the ABSOLUTE option for Shape Keys work?
things would be so much simpler if we could use the ABSOLUTE option.
just wanted you to know about me being able to use 36 shape keys (or more, haven’t tested)…
(not the stated 6+1 limit)2019-12-30 at 12:50 pm #224672019-12-30 at 2:20 pm #22468
Well let’s go back to this thread full circle…
“ARCS” AS “PATH” IN PARAMETRIC MORPHING? (3D PIE CHART RADIAL GROWTH) …
After a month using material solutions to fake it, I came back to it and fixed it.
Yes lol it’s working very well.
One object with 22 parts, 36 shape keys… Then I separated the object into its separate 22 parts which kept the 36 shapes each…
Which were changeable each.
2019-12-30 at 2:34 pm #22471
- This reply was modified 7 months, 1 week ago by GlifTek.
What’s interesting, is I went to recreate it (with more refined detail) because of the perceived bug, and that attempt rendered imperfectly.
So I went back to the saved buggy version and experimented, eventually correcting it.
I’m going to try to fudge the new version to work, using what I did to the original.2019-12-31 at 5:26 am #22477
- You must be logged in to reply to this topic.