Second Life Fitted Mesh 2013-48

Discussion of Fitted Mesh is all the rage in the Second Life blogosphere and forums. Everyone has an opinion. Ciaran Laval has an article out: Qarl – The Simplest Solution Is The Best One – And Collision Bones Are Indeed MUCH Simpler Than The Mesh Deformer. The title pretty much says it. You can read Karl’s words in the SL JIRA STORM-1716 here. He also comments in this week’s Metareality podcast, which I haven’t gotten to yet.

Quoting Karl from the JIRA:

Qarl Fizz added a comment – 20/Nov/13 1:00 PM

several people have asked me – this seems like the best place to answer.

LL’s assessment here is mostly good. in almost all situations, the simplest solution is the best one – and collision bones are indeed MUCH simpler than the mesh deformer. as i see it, collision bones have two downsides: 1) they are substantially harder to use for the person creating the garment and 2) probably don’t track as well to the avatar shape.

in the end, the evaluation must be made by the content creators who use the tool.

i will reiterate that the two year delay and refusal to communicate are unacceptable.

There are a number of opinions out that I believe reveal a high degree of ignorance and a significant degree of malevolence toward the Lab. I am certain they have no proof/evidence to support their opinions. So, take those opinions as self revealing expressions of transference and keep a distance from people assigning evil motivations to the Lab.

Oz Linden’s recent comment is:

Oz Linden added a comment – 20/Nov/13 10:38 AM

Much as I appreciate the thanks, if you really want to thank me, go download the new project viewer and don’t comment any more on this issue. [STORM-1716]

(sorry… I’m just tired of the Subject line appearing in my Inbox)

There are very few people using the Fitted Mesh Viewer. Since it is a Linden saying ‘very few’ I am not sure if that means a few hundred or a few thousand.

If you decide to install this viewer, open your currently installed viewer. In Preferences->General tab, set your viewer updates to MANUAL. Close the viewer and then install the Fitted Mesh Project Viewer. Once installed start the viewer, before logging in, open Preferences and make sure the viewer is still set to MANUAL updates. Otherwise, the viewer may revert to the main or an RC viewer… I ended up having to install the viewer twice.

If you have the viewer installed properly the HELP->About…. will show you running: Second Life 3.6.11 (283899) Nov 13 2013 20:44:54 (Second Life Project FittedMesh). The bolded part being the import part.

It Is Harder to Weight Fitted Mesh

Several people are saying it is harder to weight Fitted Mesh than Deformed Mesh. From some viewpoints they may be right. But I don’t really see it being harder. We do however have 4 new bones to weight for and I have not seen any initial weighting as there is for previously existing bones. So, there is more to do. But, either way I build a basic dress and basic shirt to use as my master source for weighting. I do a careful job of weighting those items. Then transfer those weights to new items. So, once I have a model working as I want, weighting will no longer be an issue. All future weighting will be mostly tweaking.

Having master weighted items makes the amount of work and difficulty about the same for Fitted or Deformed Mesh. So, from my perspective, they are the same.

How to Weight Collision Bones

If you are wondering how to do collision bone weighting, check the tutorial on Jump to time mark 13:25 to see just the collision bone information. It uses Avastar. But, you’ll see the process. I don’t have enough details to do a tutorial yet. I am still experimenting.

Other Viewers

Inara put up a picture of how Fitted Mesh looks to people without a Fitted Mesh capable viewer. Bizarre. So, we are not going to see third party viewers (TPV) enabling this feature in their main releases before the Lab OK’s its release.

Not only does the avatar change (and the viewer’s avatar files) but the viewer code and how it renders the avatar is changing. Those changes are in process now.  Once enough changes are added and tested, a new project viewer release will come out.

The number and amount of code change is small. So, this feature will be easy for TPV Dev’s to implement. I expect all the viewers to have it within days of or the day it is released.

4 thoughts on “Second Life Fitted Mesh 2013-48

  1. “Several people are saying it is harder to weight Fitted Mesh than Deformed Mesh. From some viewpoints they may be right. ”

    To me its really a cake vs. better cake argument compared to 3-4 years ago when I was creating and attaching folders full of sculpties.

    I don’t think there’s a “may” about it though, having to weigh for bone scale and movement (in cases like cleavage) in addition to rotation is added complexity. It’s the downside.

    The new solution is simpler for Linden Lab as Qarl states, but definitely not simpler for creators when compared to the mesh deformer. I hope improvements can still happen in the future. It’d be nice if for example it was technically possible to provide two sets of weights; one for the normal animation bones for rotation purposes and one for the collision bones for scaling purposes.

    I think this first pass shouldn’t be complicated by feature suggestions though, we’ve seen how those could distract in the deformer project. This solution is good enough and much welcomed. It’ll shrink product folders down to one mesh and one alpha in most cases and eliminate standard sizing (maybe, the split concern in weighing might make the extreme shapes impossible).

    But I still wish we got Qarl’s solution. The products would be better, and they’d be easier to make. So hopefully this doesn’t close the window on further improvements, or alternatives down the line.

    • I mostly agree with you. This is a good step forward.

      I am not sure Karl’s Deformer would be that good overall. But, it was excellent for mid-range avatar sizes. I think Fitted offers many more possibilities.

      Feature requests may confuse things. But, we either ask for them now or do without. Plus, these suggestions and experiments are going to reveal if this is a mistake or not. Or they may lead to new ideas. So, I encourage people to shoot for the moon.

      This is different than in the Deformer scenario. There the Lab realized there was a lot of fixing to be done and worried that the fixes would lead to breaking the Deformer. It was pointed out that months before Redpoly came out with the Collision Bone weighting Prep Linden was asking for examples of collision bone weighting. That suggests to me, that the Lab was already looking at the mesh clothes fit problem. Now that they have almost 3 years of thinking on the problem, I suspect they have a plan for how the avatar will develop.

      I found it interesting that as slow as things seem to be right now, the Lab had no people to spare for LEAP Motion integration.

  2. I’ve not downloaded the “Fitted Mesh” project viewer yet, since I am looking at this as a customer waiting for the market to resolve. Is using that viewer a way to show ‘support’ for this? Or is there enough noise already that LL knows this is a major issue?

    • I think the more people using the viewer the better. As it is a separate install independent of the main viewer or RC’s, it is an easy non-interfering install. But, the project viewer tends to catch up and lag behind the main viewer. So, it is not the most pleasant viewer to use, the Interesting RC is way better. Also, unless you plan to upload rigged mesh, I’m not sure whether just use of the viewer will impress the Lindens.

Leave a Reply

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