There was a Third Party Developers’ meeting this past week. Of course Project Bento was a main topic and Vir Linden was there to talk about and respond to comments in the Feedback thread of the forum. So, what did we learn?
Viewers
The Second Life Valhalla Viewer version 4.0.0.307894 was promoted to default viweer status Thursday 12/17. This viewer has fixes for Bento, which mean little as Bemto animations and avatars are not yet allowed on the main grid. It also changes from Webkit to CEF – Chromium Embedded framework.
Co-Routines or Second Life HTTP Update Viewer version 3.8.7.308134 is merging with the Second Life Vivox Viewer version 3.8.7.307744 and will reappear in RC as a single viewer named… we don’t know.
The Second Life Quick Graphics Viewer version 3.8.7.306758 will get the Valhalla updates.
Next viewer promoted is likely to be the Second Life Maintenance Viewer version 3.8.7.308556. The Lab’s preference is the HTTP viewer promote. But, testing decides and apparently this one is performing better.
Oz Says he is getting someone back on the Second Life Project Oculus Rift Viewer version 3.7.18.295296 (TM = 5:00). In my review of 2015 I point out this is something they have been looking at doing all year. As VR is nearing and many are predicting a consumer release in first quarter 2016, the Lab has likely waited as long as they can. So, I expect to see an update in the near future.
Bento Viewer
Currently: Second Life Project Bento Viewer version 5.0.0.309171.
The Bento Avatar is crashing the Firestorm Viewer, poof’n to desktop as soon as a a Bento Avatar comes into sight. Oz Linden tells us that for the past few Linden Viewer releases they have been adding code to fix that problem. You can see the viewers that have the fixes in them as the Bento Avatar appears as a skewed avatar. The crash has something to do with the weighting values and division by zero. The skew somehow works around that problem.
You may be able to imagine the chaos that could be caused by griefers running through Second Life events wearing a Bento Angle avatar. The Lab has quietly been building in a fix for that problem over the last few viewer releases. Version 4.0 has the fixes and is, AFAIK or anyone knows, immune to Bento crashes.
The crashes are mostly on Firestorm, possibly ONLY on Firestorm. This concerns Jessica, FS Team manager, as the team will need to get a release out before Beto Avatars start appearing on the main grid.
Oz, and Bazz (or Vir Linden) figure that any viewers updated to the Linden 4.0 code won’t be crashed by Bento. The fixes for Bento crashes are in 4.0 and on. Oz figures that Bento won’t make it to the main grid before the end of January 2016. If things go well, it might make it out in February and possibly farther out. But, the Lindens are trying for as soon as possible.
The next sort of on schedule release for Firestorm would be February. So, it looks like Firestorm users are going to be well taken care of, but they will be updating and older versions will have problems.
27:00 Vir starts to talk about BUG/Feature 10990. Inara has clipped that bit of audio out of the TPV Dev’s meeting video. It seems moving bones is a problem for the viewer code. So, the Lab is hoping to avoid allowing bone position changes. They did consider animating bone positions and Vir addresses the forum comments about needing to position bones for facial expressions.
The problem that decided them against it is avatar scaling… size changing. It seems the thought is that changing the size of your avatar from the size the animation was designed for would distort the animation of the face or body.
I see something and suspect you have too that is representative of the problem in a number of animations that I/we have. Some animation vendors sell 3 versions of their animations, one each for; small, medium, and large avatars. I like those as they do look better on my avatar. The ones I make for my avatar are near perfect. But, I use a near exact copy of my avatar in Blender. If I animate a hand resting on my leg, it actually looks like it is on my leg. But, that isn’t so for avatars that aren’t 5’-10” and have different length arms than mine. Animations in SL don’t scale. I don’t sell my animations. I’ve thought about it.
30:06 – Oz makes his suggestions for what can change the Linden thinking and likely lead to consensus. Assertions that ‘A’ can’t be done with present restrictions or ‘B’ won’t work well without ‘C’ are not helpful. More concrete would be providing Animation #1 showing problem A and providing mesh B with its animation showing how poorly it works without ‘C’.
Without examples and tested user cases with proofs it is unlikely the Lindens will be convinced. This is like trying to prove innocence after a conviction. It takes exceptional effort. If we want something other than what the Lab is currently willing to provide, the burden is on us. Don’t talk to me about fair. Grow up, life isn’t.
Oz’s belief is this will move the conversation into a demos and fact based discussion of what can and can’t be done. I think his hope is those that know how to accomplish the tasks some fall short of will come forward. The result being all will see what is and isn’t possible. To me it seems the Linden minds are open. But, it will take solid thinking and demonstrations to move their thinking. I STRONGLY suggest you click the 30:06 link above and listen to Oz’ well said request. I think he will add a similar statement to the JIRA. But, in the video you can hear his thinking coalescing.
The Lindens listen, in spite of what many say. If you say it nicely, they are more likely to pay attention. Well, that is basic human nature…
Another page… link below…