I've had more troubles again.  While trying to rotate those robot model fingers
to open and close they tend to jump to the opposite side of the intended
rotation.  Of course that means I'm not using IK all the time but I wouldn't
have thought it to be a problem when switching to translate rotate scale.
And even though I'm now getting the correct behavior of the keyframing there's
still some oddities about the cubic interpolation, I just don't know if it's
also correct and I'm just expecting something else.
I had moved the block away from the robot hand too by cut and paste (wish I knew
of a way to retain the original coordinates as well) so as to remove it from the
robot group and place it on the floor by itself.  Then later cut and paste back
into the hand so it's grouped with the robot arm once again after the picking-up
part of the animation.
All in all still a lot for me to learn about all of this.  Hope Lutz &
Kretzschmar don't have a media explosion animation in one of those nag boxes
:-)  I think my 75.25 US$ is going to be on it's way soon anyhow though.

Bob

"Mikael Carneholm" <sa9### [at] idautbhbse> wrote in message
news:38ABC1F5.7C6160CE@ida.utb.hb.se...
| Bob Hughes wrote:
|
| > I guess that's 'linear' and not 'cubic' interpolation.  I found already it
can
| > be tough to control 'cubic'.
|
| Oh...sorry...I seem to have missed that it all was about cubic interpolation.
| Anyway, I have used cubic interpolation in a "sight-seeing" trip though my
| apartment - it works great. Too bad the animation is 40 Mb, otherwise I would
have
| posted it...
|
| ----------------------------------------------------
| Mikael Carneholm
| Dep. of Computer Science and Business Administration

|
| Personal home page:
| http://www.studenter.hb.se/~arch
| E-mail:
| sa9### [at] idautbhbse
|
|

Post a reply to this message