The following notes and audio were taken from the weekly Bento User Group meeting, held on Thursday, July 14th at 13:00 SLT at the Bento User Group wiki page.. For details on the meeting agenda, please refer to the
Note that this update is not intended to offer a full transcript of the meeting, nor does it present the discussion points in chronological order. Rather, it represents the core points of discussion, grouped together by subject matter were relevant / possible, whilst maintaining the overall context of the meeting.
The Bento project viewer updated shortly ahead of the meeting to version 126.96.36.1997597. This is intended to be the final round of skeleton and slider changes, and as such includes the modifications to the head bones that were previously offered through the experimental viewer, which can be summarised as:
- A new bone, mFaceJawShaper, intended to resolve issues around the jaw angle (see my Bento update #17 for background)
- Additional tweaks to existing facial bones
- Some slider changes.
In addition, the viewer includes the avatar height update referred to in my Bento update #18, although this has been causing some issues, and some quasi-scaling for wings and tails with avatar height.
These latest updates may mean that content built using earlier versions of the skeleton may no longer display correctly in the updated viewer.
With this release, the Lab is not intended to make any further significant changes to the skeleton or sliders, but focus solely on bug fixes and moving the viewer to Release Candidate status. The only exceptions will be if major regressions are found which affect pre-Bento content, or major issues of a similar nature. If such issues are found with the project viewer, the request is to let the Lab know via bug report and forum thread feedback.
There are also updated avatar test models to go with the updated viewer, which can be found on the SL wiki Bento testing page. These also include mesh eyes now that scaling has been added to the eyes.
The hope is that allowing for vacations and there being no significant issues found within the July 14th project viewer, it will move to RC status in August. This may also be dependent on whether or not the Lab an resolve the issue of facial distortions occurring at altitude when software skinning is used, which is still being examined.
Should finding a resolution to this issue continue to prove problematic, and providing the issue does not cause performance issues, the Lab might leave it unresolved with the recommendation that creators use hardware skinning, so that the viewer can be promoted to RC status with this flagged as a known issue.
Any progress from RC to release status for the viewer will depend on whether further bugs or issues are found within the Bento viewer, what other RC viewers are in the release channel, what the relative priorities are for the viewers in the release channel in terms of promotion to release status, etc.
Questions were raised in the meeting about making use of the body fat slider with Bento to make for easier weighting of body weights with mesh heads. However, the body fat slider works entirely on morphs, so to be effective with Bento, it would require the addition of additional body joints, which is not on the cards at this point in Bento.
Following the discussion about the Alt_Eye bones, some modifications have been made. In particular, they are now parented to the face root bone, rather than having the same parenting as the original eyes, although they still respond to the same sliders. The viewer has also been updated so that it will animate both sets of eyes consistently (e.g. looking at speakers or other visual targets).
The re-parenting of the alt_eyes is intended to make things eyes for people wishing to use them on non-human avatars. This isn’t as general-purpose as had been hoped (see my update #18), but there were constraints on how the original eyes could be handled for non-human avatars without risk of breaking pre-Bento content, so this route was seen as the best means of presenting people with alternative eye sets, rather than completely re-purposing the alt-eyes.
Rigging to Attachment Points
During the development of Bento, the Lab were introducing validation checks on the simulator side to improve reliability in SL (e.g. checks to ensure animations are what they say they are, further checking of mesh uploads to ensure they can be rendered correctly, etc.). A side effect of these updates would have been that it would no longer be possible to upload mesh items rigged to attach points.
This caused a degree of consternation among content creators (see my SL project update for 2015 week #44), and the decision has now been made not to disable the upload to content weighted to attachment points so as to prevent breakage with existing content (e.g. when issues updates). However, the Lab will not be encouraging people to rig to attachment points.
This relates to BUG-20013, and as reported in my update #18, linked to above. In short, an avatar’s position relative to the ground it is standing on is calculated using the position of a number of bones within the avatar and various other factors. However, how and when these calculation is updated within the viewer occurs in a somewhat unpredictable manner, with the result that while an avatar may appear to be standing on the ground in one view, it is standing in the ground when seen by others.
To try to correct this, Vir has updated the viewer so that the recalculation of an avatar’s position relative to the ground is made more-or-less constantly. This should mean that every viewing a given avatar should see it positioned correctly, relative to the ground. Unfortunately, this has resulted in some unintended side effects. For example, and as reported by Aki Shichiroji in BUG-20169, if an animation is acting on the position of the bones bones used in the calculations of the avatar’s height / position, it’s impact can lead to some odd results, as Aki demonstrates via two .GIF files.
The movement is also translated to the avatar’s camera, causing it to move up and down, causing visual discomfort for the user.
It isn’t clear what can be done about this, although Vir will look into the issue. In the meantime, the recommendation is that people avoid changing the position of any of the bones used in the position calculation via an animation, and instead change the rotation of the bone(s) to achieve the desired result, and see if that eliminates the problem.
There was some concern / confusion in the meeting as to whether this would in fact work within the tools, which sparked a further discussion, culminating in the suggestion that unwanted translations could be removed / muted in Blender prior to export to tools like Avastar, which may also correct the problem.
Matrice has requested that issues involving the export of animations from Avastar should be reported to him or Gaia, so that they can investigate issues more deeply.
Applying Joint Position Adjustments
The discussion on avatar height above ground led to clarification being required on the order in which joint position adjustments are applied, with Vir noting that it should be in the order: skeleton initialisation, then sliders, then effects joint overrides then effects of any animations, and this should be globally consistent. This is also the order in which the Reset Skeleton option applies joint position adjustments. He requested that if this is not the case, that JIRAs outlining conflicts / different ordering are raised.
Vir is on vacation for the next two weeks. However, the Bento meetings will be held on an informal basis to allow content creators to discuss the viewer updates, issues, etc., and Coyot Linden will be available to relay comments and feedback to the Lab.