Discussing Winged Edges and a better Skinning (copied here)

The interface, modeling, 3d editing tools, import/export, feature requests, etc

Moderators: jesterKing, stiv

thorax
Posts: 320
Joined: Sun Oct 27, 2002 6:45 am
Contact:

Discussing Winged Edges and a better Skinning (copied here)

Postby thorax » Sun Jun 22, 2003 9:38 am

I'm going to continue my descriptions of my
ideas for skinning and Winged edges at
my website at

http://www.bl3nder.com/UML/WingedEdge.html
http://www.bl3nder.com/UML/SkinningAndIK.html

I plan to do visual descriptions of these in addition to the
UML.. The discussion on Skinning models
my experiences with IK and skinning on Maya, Alias Poweranimator
and Kinemation, from what I can recall..

For the math majors..
Alias Poweranimator's IK + Wavefront's Kinemation + Some Constant = Maya's IK..

If anyone has any contrary or additional ideas based on
these, they should either suggest them at the functionality
board, here, or send me email at rofthorax@canoemail.com
and I'll include them in my notes.. You can access anything I'm
designing programmatically at www.bl3nder.com/UML .

I removed my blog for reasons of being fair to Ton and others in
not bad mouthing them somewhere else.. I will try to be more
descriptive and idea oriented.. The work on the UML is a
step toward trying to make my ideas visual and understandable..
The skinning system and indeas about winged edges
will be more apparent when I made visual representations
of the ideas in relation to blender's interface. But these UML
diagrams are more for the logical/structural relationships
that the programmers would consider. However the UML diagrams
don't represent any particular details of implementation or
of any particular language, though a class diagram implies C++
and Java a lot I'm talking about functional and data relationships
in these diagrams.

The reference I made to Ton's implementation of IK
was based on the old implementation of IK, I don't know what
is used now.. And I'm not sure if Ton does either since at NaN
I had heard Ton was dealing indirectly with the coding and
other people were dealing with it. Am I wrong here?

My assumptions that the IK and skinning are not optimal are
based on my experiences with high level packages and
using slower machines as compared to using blender on
similar or faster machines. The users are making due with
skinning of Subdivision surfaces to bones, which is
probably a good solution, but is not optimal as
any IK implementation on a commercial package such as
Maya, even in realtime video games, will
do deformation and IK in relatime on meshes containing
thousands of vertices. Its not a good argument for blender
and should be recognized as a lacking, functionally.
But the hacks that allow users to make do with less is
acceptable temporarily.. Blender users have been forced to
deal with less, not because its all that is possible
but because nothing is said about its lacking as the more experienced
users avoid raising such fusses in the group for fear that
it might detract from the acceptance of blender.. Blender is
fantastic and straightforward, its excellent for modelling and
basic animation, but character animation its still lacking some
luster at.. The relative vertex keying is probably the best
character animation feature, and the IK is functional but clumsy to
control. The lattices are great as deformers but would be better
used as deformers of the cluster concept that I introduce in the
page about skinning..

Some have said that blender lacks nothing really, either
are using blender with other software that makes up for what it
is lacking in, are using techniques that allow them to do what they want,
or are not aware of what better technologies exist. A sit stands blender
is great as a utility to some other 3D package or as a fixit
application for what can't be done in another application.

I saw elsewhere Eskil mentioning an idea about whether blender
to replicate features onf Wings, and said that that would be cloning..
What I wonder is what is wrong with cloning of functionality of a
lesser application, which Wings is.. It lacks the functionality and
navigation that make blender great, that's why blender should
absorb the features of Wings.. Eskil makes the point probably
in relation to Verse which is supposed to act as a glue between
client applications, possibly using the best of every application,
which could be a form of reuse.. But still I would like Wings
modelling in blender.. IT could be possible to access Wings
modeller through blender via Verse. And this wouldn't
put a strain on blender's working.. But I have other issues
with Verse that I need not repeat here.

Cloning is not a bad thing if blender reuses the source of
Wings, which it could.. It may be the case that Wings
becomes a library anyhow, and the client interface was just to
encourage others to consider and reuse the concept of Wings.

Now I tend to read a lot into concepts presented by others,
don't take it personally.. I'm just trying to make sense of it
and relate it in a way that seems meaninful to me.. And divulging all
for othrs to consider and cognate.. If you disagree, tell me
why.. This is the point of discussing blender's design..
Not the functionality, but the design..

matt_e
Posts: 898
Joined: Mon Oct 14, 2002 4:32 am
Location: Sydney, Australia
Contact:

Postby matt_e » Sun Jun 22, 2003 10:17 am

(Offtopic:
Thanks for re-posting in the appropriate forum. I've removed the thread from News & Chat.)


Return to “Interface & Tools”

Who is online

Users browsing this forum: Yahoo [Bot] and 1 guest