[opensource-dev] Content Creation Improvement UG - Morph Target Discussion

Geenz Spad geenz at exodusviewer.com
Mon Jul 16 13:33:35 PDT 2012


Hello everyone!

Recently at the Content Creation Improvement meetings, we've been discussing morph targets and what they would mean for content creators on the grid.  Thus far the conversation's focused on how this could benefit clothing designers; e.g., what level of control would this give them, and why it would be beneficial for them to have this feature.

Thus far, there hasn't been too much developer feedback on the feature.  Siana Gearz has of course posted some possibilities in the comments on http://blog.nalates.net/2012/07/10/the-consensus/, and does have a valid point that there are some technical hurdles that will need to be overcome with whatever approach that is taken.

Before I get into what should be discussed, a little definition of what a morph target is:
A morph target, also known as a blend shape or shape key in some 3D modeling packages, is something that defines how the surface of an object should "deform" when a value is applied.  You could say that the Second Life avatar mesh makes use of morph targets to define the overall shape of an avatar as an example of a morph target.  Video games now days tend to use them to define facial animation, and other animated properties that would just be far too difficult to animate using bones.  In this context, we're primarily looking at them to define how rigged meshes deform, however, that being said, other general applications being discussed wouldn't be bad either.  But please focus on how we can improve clothing design with such a feature as a primary goal, with other applications being a secondary goal.

I'm extending the discussion to the open source dev mailing list, in the hopes to get more developer discussion going on the matter.

Currently, there are a few things that need to be established through these discussions:
• What can morph targets do that existing tools can't (Qarl's deformer included)
• What problems can they solve that existing tools can't, or otherwise are difficult to solve with existing tools (Qarl's deformer included)
• What can they generally be used for, from various view points
• Ways they can be implemented
• Ways the bandwidth cost can be mitigated for those implementations

Just as a note with regards to Qarl's deformer, and where this may eventually go:
Qarl's deformer comes first.  Thus far, it is the closest solution to making mesh clothing fit better, and it requires very little work on the content creator's behalf outside of modeling around the default "Ruth" shape.  As such, morph targets are not intended to be a competitor to the mesh deformer, it's intended moreso to complement what the mesh deformer is capable of doing as a more advanced toolset in the context of clothing.

Qarl's deformer has the highest likelihood of being released before the end of the year in a final form, and morph targets may not even be seriously considered by the lab until sometime next year.  On top of that, having both would have benefits for content creators; some content creators may not want to learn how to use morph targets and are content with modeling clothing around Ruth, and many may want to learn how to use morph targets "later on" while they focus on more immediate results in the interim with the intention of further customizing how their clothing deforms later.  So really, this should be seen as more of an advanced tool for content creators that wish to have further control over the final outcome of the clothing they design.

These discussions will be brought up at the CCI meetings over the coming weeks, and it's my hope to see more developers who participate on the list pop in and contribute things to the meetings, or even just sit in and watch the discussion, as I'm sure there's other things being discussed that would be of interest to developers being brought up at the meetings.

An additional note is necessary here: these discussions are informal.  Because we discuss this, does not mean it will be implemented by the lab.  However, we can organize things in such a way to produce a proposal to the lab, and (eventually) prototypes for consideration with LL's permission once a clear vision has been established for the feature, and its associated goals.  From there, it'll be up to LL to decide if a feature is worth having, and allotting the resources to make these things possible as needed on their side potentially with open source developers contributing to the final viewer that will make use of these features.  The hope is, this will increase the chances of new functionality and other improvements regarding content creation being added to Second Life.

You can find the agenda for the meetings, and other information relating to them, here:
https://wiki.secondlife.com/w/index.php?title=Content_Creation_Improvement_Informal_User_Group
--  
Geenz Spad
Sent with Sparrow (http://www.sparrowmailapp.com/?sig)



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.secondlife.com/pipermail/opensource-dev/attachments/20120716/060b9157/attachment.htm 


More information about the opensource-dev mailing list