Good Solutions

This week we discuss the status of the Second Life mesh deformer, Linden Lab’s new fitted mesh solution, Bitcoin, Google Glass and more!

This week’s topics include:

Guests:

Karl Stiefvater (Qarl Fizz, formerly Qarl Linden) is visual effects artist, software developer, and interactive designer extraordinaire.

William Reed Seal-Foss is a 3D artist and virtual world content creator.

Thank you to our sponsors – Botgirl’s Digital PlaygroundAngelRED CoutureKahruvel Design and Virtual Vision 2020

About Kim/Gianna

Kimberly Winnington (SL: Gianna Borgnine) is the Emmy nominated owner and CEO of Sand Castle Studios, LLC, a company dedicated to helping organizations maximize the full potential of virtual worlds and social media by creating interactive, social, and 3D experiences. For the last 5 years, Kimberly has helped SCS clients stay ahead of the curve and is a respected resource for information on current and developing trends, social media, and immersive experiences.
This entry was posted in Podcast and tagged , , , , , , , , , , . Bookmark the permalink.

10 Responses to Good Solutions

  1. Pingback: Qarl – The Simplest Solution Is The Best One – And Collision Bones Are Indeed MUCH Simpler Than The Mesh Deformer » Ciaran Laval

  2. Cathy Foil says:

    Great show as always! :D

    I am so disappointing that the Mesh Deformer has been rejected in favor of the Collision Bone method. I have extensive experience with both methods and I can tell you from a content creators perspective the Mesh Deformer is way less work and the deformations to the mesh clothing are better.

    The Collision Bone deformer works OK for loose fitting clothes but really is lacking when it comes to tight fitting clothes. You get way more skin poking through. Because the Collision Bones are offset from the normal bones mesh does not bend in the same places as mesh weighted to the normal skeleton which makes even more skin poke through when an avatar moves.

    Other reasons why the current Collision Bones are difficult to work with are that their rotations are not set to zero and the default size scale settings are not X=1, Y=1 and Z=1.

    Adjusting the weighting to the Collision Bones are much much more difficult than adjusting the weighting to the normal skeleton bones because now you don’t just have to worry about how the mesh moves when the joints bend but also how smoothly the transition areas are between multiple joints as the scales of the joints change. I guess it is 3 to 5 times the amount of work to get the weighting right.

    Now some of this can be made a lot easier if LL would move the Collision Bones so they are in the same exact positions of the normal skeleton and made the rotations zero and scales all set to 1 for the default Ruth settings but you still have the weighting adjustment difficulties. So maybe LL could help us get the time down to twice as long?

    I really wanted the Mesh Deformer because with it we could have replaced body parts of the default avatar and have them attach seamlessly. Not only seamlessly but also things like mesh hands and mesh heads would move and faces have facial expressions because they would be driven by default avatar mesh through the Mesh Deformer. This is something the Collision Bones just can’t do.

    We the residents could have created new complete mesh avatars that worked with every animation override and that had full facial expression and hands who’s fingers moved without having to do anything other than clicking two options in the upload window to make use the Mesh Deformer.

    We wouldn’t have wait for LL to create Ruth 2.0. We could have made out own. Good luck getting LL to let us upload morphs. :(

    Oh I did want to let you know something interesting I found out. Using Collision Bones was not RedPolly’s idea. Well at least he/she wasn’t the first and I don’t know if someone at LL gave RedPolly the idea or push to do it. Here’s a link to a post by Prep Linden asking for volunteers to weight mesh to the Collision Bones more than 7 months before RedPolly came out with their video. http://community.secondlife.com/t5/Mesh/Need-some-samples-of-a-rig-weighted-to-the-collision-volumes/td-p/1142973

    I posted a plea to the community to support a solution I came up with where we could have the Collision Bone deformer as the default and the Mesh Deformer as a graphics option. I detailed how the solution could work. Please go there and add your support.
    http://community.secondlife.com/t5/Mesh/Solution-We-Can-ALL-Be-Happy-With/td-p/2341689

  3. Ciaran Laval says:

    I’ve never looked at either solution but from reading the Jira on the deformer it seems to me that both solutions could have worked side by side as one of the intentions of the deformer was to have a boolean value to say whether the mesh item could be deformed or not.

    This was to ensure that already rigged items wouldn’t be deformed and to ensure backwards compatibility but wouldn’t that value also have worked to prevent fitted mesh items from being deformed by the deformer too?

    • Cathy Foil says:

      Ciaran Wrote:
      “This was to ensure that already rigged items wouldn’t be deformed and to ensure backwards compatibility but wouldn’t that value also have worked to prevent fitted mesh items from being deformed by the deformer too?”

      The answer is yes and no. Fitted Mesh clothing, that is mesh weighted to the Collision Bones also known as Liquid Mesh, uploads like any other mesh that has been weighted to the normal skeleton bones. It is up to the content creator when uploading the mesh whether or not they want to select the options for making the Fitted Mesh also mesh that can be deformed by the Mesh Deformer. If a content creator just wants their mesh to be deformed by the Fitted Mesh Deformer they can by just not selecting the Mesh Deformer option in the upload window. Also the opposite is true. Clothing mesh weighted to just the normal skeleton bones and not the collision bones are not affected by the Fitted Mesh Deformer.

      So mesh content creators would have four options:
      1. Mesh that can be deformed by the Fitted Mesh Deformer only.
      2. Mesh that can be deformed by the both the Mesh Deformer and Fitted Mesh Deformer.
      3. Mesh that can be deformed by the Mesh Deformer only.
      4. Mesh that is NOT deformed by either Deformer.

      The smart clothing designer will pick option two.

      Both deformers can work at the same time on the same mesh. I know I tried it. What you get is a slightly different deformation than either alone. I suggested a boonlean be added so that when someone in their graphics preferences turns on the Mesh Deformer that it turns off the XYZ scaling of the Collision Bones basically turning off the Fitted Mesh Deformer. Or perhaps this should also be a separate option. Or perhaps turning on or off either deformers be an option one that just affects their avatar. This would make it possible for each resident to choose which deformer they want others to see. There are all sorts of ways this can be done.

      What upsets me the most is now knowing that LL never had any intention of using the Mesh Deformer all because Karl wrote the code. I believe him when he says he has heard from two inside sources that confirmed this. This means I have wasted probably about 20 hours of my life testing the Mesh Deformer and hundreds if not thousands of hours as the only reason I stayed active in SL was the promise of the Mesh Deformer being adopted eventually. I waited patiently for over two years and now I find out it was a lie!?

      My time in SL is done. The change in the TOS and now this! I won’t delete my account but I am going to cancel my premium membership and sell all my land. I am going to move to InWorldz and possibly Cloud Party and make things there. I can get a lot more land for my money there and make the things I had planned to for the Mesh Deformer.

      My posting in the Mesh Forums where I plea for implementing both mesh deformers is a hail mary. I doubt LL will actually listen or implement it.

  4. Cathy Foil says:

    Karl this is an InWorldz question.

    Would it be possible to create an attachment point on the avatar specifically for custom mesh avatars that once attached to this special point would have access to all the normal skin and clothing layers of that the avatar is wearing?

    This way what ever textures and layers that you were wearing would automatically be applied to the custom avatar mesh or even mesh that was replacing just part of the body. Minus the Alpha layer for the avatar of course.

    • qarl says:

      Cathy – that’s a fascinating idea. i’d implement it differently – with a tag on the uploaded mesh instead of an attachment point.

      regardless – applying the standard avatar layers to a customized avatar mesh is brilliant.

      technically – you wouldn’t be able to use all the individual clothing layers – just the final composite. but that’s enough to let you “upgrade” the LL avatar shape without getting LL to do it themselves.

      i love it.

      • Cathy Foil says:

        Hey Qarl antoher InWorldz theoretical question.
        The reason I am putting forth these questions and or ideas is I am hoping to spark a dialog not only with you but with anyone who has a desire to improve InWorldz and Open Sims in general.
        I address these idea or questions to you because you are the best person I know to evaluate whether or not an idea is possible or practical.
        I am not one who has to have their idea be the best or the one that is used.
        I love it when my idea sparks an even better idea in someone else.

        My hope would be to get a set of improvements and innovations that perhaps the owners of InWorldz would fund you or someone else to do.
        If they don’t have the financial resources to do them perhaps a Kickstarter can be done and then the results shared with all Open Sims.

        OK back to my original question. The Mesh Deformer is like 95% of awesome. The other 5% is not due to Mesh Deformer but the deficiencies of the default avatar mesh itself and its horrible blend or morph shapes. We all know a better new avatar with better morphs would fix this. My question is this instead of using the actual default avatar mesh to deform could a second avatar mesh, that is invisible, that has all its own blend shapes and or morphs, be used instead?

        See what I picture is a second invisible avatar mesh which has more vertices more evenly spaced, better weighted and who’s blend shapes or morphs closely match the original default avatars blend shapes or morphs.

        Not all avatars would have to have the second invisible avatar mesh. It would automatically be worn when someone wears Mesh Deformer mesh.

  5. Sean Brady says:

    Please keep the uncensored version. Less work for you, and less bleeps for me. I know what Reed is saying even when you bleep him. As for the deformer issue, I am glad to have a solution in the works. My personal opinion is that while Karl may have heard something from LL no one knows why the decision was or was not made. Maybe they didn’t ever plan to release it because they simple thought it was the wrong method, maybe it was just spite. Karl is right though in saying that the deformer did have the affect of keeping discussions active and in the mind of developers. We have a solution, and I think that is a win for everyone.

    In closing…yea..I wish I had paid a little more attention to Karl’s earlier praises of bitcoin. Sigh.

  6. Inara Pey says:

    A second attempt to post this…

    @Cathy

    “Oh I did want to let you know something interesting I found out. Using Collision Bones was not RedPolly’s idea. Well at least he/she wasn’t the first and I don’t know if someone at LL gave RedPolly the idea or push to do it. Here’s a link to a post by Prep Linden asking for volunteers to weight mesh to the Collision Bones more than 7 months before RedPolly came out with their video. http://community.secondlife.com/t5/Mesh/Need-some-samples-of-a-rig-weighted-to-the-collision-volumes/td-p/1142973 – See more at: http://metarealitypodcast.com/good-solutions#sthash.7ezgLu3s.dpuf

    According to some involved in the closed mesh beta in 2011, Prep’s request came about following a conversation with RedPoly in which the subject was initially mooted, possibly by RedPoly. There was some discussion on this on SLU at the time RedPoly relaunched the idea in 2012.

    I’d also point-out that Oz Linden stated to me that it was RedPoly who got the Lab started on the idea when I was offered the chance to preview the Fitted Mesh viewer prior to release (see: Lab looks to make mesh garments fit better with the Fitted Mesh project viewer) & RedPoly was also present for a part of the time I was playing with the viewer.

    The thing which is of interest to me is not so much who came up with the idea per se, but rather why no more was heard from the Lab on the subject when the request for samples first went out in 2011.

  7. Pingback: The Mesh Deformer is dead, long live Fitted Mesh | Mona Eberhardt

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>