2024 week #5: SL CCUG meeting summary

Hillvale Beach, December 2023 – blog post

The following notes were taken from my audio recording and chat log transcript of the Content Creators User Group (CCUG) meeting held on Thursday, February 1st, 2024.

  • The CCUG meeting is for discussion of work related to content creation in Second Life, including current and upcoming LL projects, and encompasses requests or comments from the community, together with viewer development work.
  • As a rule, these meetings are:
    • Held in-world at the Hippotropolis Campsite, where it is chaired by Vir Linden, in accordance with the dates and times given in the SL Public Calendar.
    • Conducted in a mix of voice and text.
    • Open to all with an interest in content creation.
  • The notes herein are a summary of topics discussed and are not intended to be a full transcript.

Official Viewers Status

  • The Maintenance W RC viewer – bug and crash fixes – updated to version 7.1.3.7701974306 on January 31st, January 31, 2024.

The remaining official viewers stood unchanged from the start of the week at the time of the meeting:

  • Release viewer: version 7.1.2.7215179142, formerly the glTF PBR Materials Maintenance RC, issued December 15, promoted January 8, 2024 – numerous bug fixes and improvements.
  • Release channel cohorts:
    • Emoji RC viewer, version 7.1.3.7453691714, January 22, 2024.
    • glTF PBR Materials Maintenance-2 RC viewer, version 7.1.3.7467259489, issued January 12, 2024.
    • Maintenance X RC, version 7.1.1.7088410646, December 7 – usability improvements.
    • Maintenance Y RC, version 6.6.17.6935642049, issued November 21 – My Outfits folder improvements; ability to remove entries from landmark history.
  • Project viewers:

General Viewer Notes

  • The Emoji viewer is still on the path towards an RC update, and it is still hoped it will get promoted to de facto release status Soon™.

Graphics / glTF

  • The “glTF featurettes” work is progressing, with the viewer following the current glTF Maintenance-2 RC viewer set to include:
    • Geenz Linden’s work on Hero reflection probes and reflections / mirrors and Cosmic Linden’s work on applying glTF materials to terrain.
    • Updates which will allow the glTF / PBR swatch boxes in the Build / Edit floater display previews of the materials being selected, rather than grey boxes.
    • Possibly, the first series of additional glTF extensions – Index of Refraction.
    • These capabilities (together with 2K textures support – see below) will be sitting behind debug settings, and so a) will need to be enabled by those wishing to test them; b) should be understood to be incomplete features subject to update and change / possible breakage.
  • 2K Textures:
    • The Lab is starting to assess the potential to allow 2K textures within SL. Such support will require “careful assessment” in terms of VRAM footprint, the risk of out of memory crashes on older systems and the like.
    • Runitai Linden believes that if there are issues with the viewer attempting to load a full 2K texture even when it is very small (or very distant), then it is a case of making the viewer “smarter” it terms of how it handles and samples / loads textures.
    • Support for 2K textures is available for general testing on Aditi (the Beta grid) within the RumpusRoom 2048 region.
    • One identified win is seen as terrain texturing, as 2K terrain textures at the region level should not overly tax systems and eliminate the current issues of blurriness / low texel density.

Mirrors

  • Work has been continuing of late, with what is described as a “significant” increase in reflection quality, and is starting to look “correct” as well as working with PBR.
  • Further performance tuning is required, and the clipping is “off in a lot of circumstances”.
  • The focus is on presenting planar reflections (utilising a cub map generated by a “Hero” reflection probe which is selection on the basis of the avatar’s proximity to it), with Geenz Linden noting it has the potential to allow more “fancy things” down the road.
  • It was noted that the approach taken has to offer a balance between intended use (to create actual mirror surfaces) against unintended uses (such as using the mirror capability within a highly polished floor), rather than trying to enforce some nebulous rule of “this functionality is ONLY for mirrors”), as well as taking into the consideration that it needs to be relatively easy to use.

Terrain

  • Cosmic Linden has indicated that once the PBR shader, etc., is working for terrain as per her current project, consideration is being given to allowing the same terrain shader to in-world objects.
  • So, for example, a mesh tunnel might be created and placed in-world, and then the outer face of the model set to display whatever the surrounding terrain might be.

“Combat Committee”

A reminder that:

  • The first meeting of the “Combat Committee” User Group will be held on Thursday, February 8th, at 13:00 SLT, and so will be alternating with the Content Creation User Group.
  • The venue for the meetings will be:  Longfellow/142/255/30.
  • The initial format for the meetings will be text-only, but this may become a mix of voice / text according to attendees’ preferences.
  • Further details are available on the SL wiki, and the meeting is now on the SL Public Calendar.

Rider Linden acknowledged the fact that I (aka “someone”) had pointed out having two user group meetings with the same abbreviation (CCUG) alternating between the same time slot on the same weekday might be confusing for some (even if the meetings are in different locations), so the name for this meeting might be changing.

In Brief

  • The transition from Jira to Github Issues  / Canny is still in progress at LL and is taking up resources and time, hence the slow-down in some areas of work (such as RC viewer updates. As per previous reporting, the work is due to be completed by the end of February, when Jira will cease to be used.
    • The use of Canny remains a source of unhappiness among those who tend to raise bug reports, largely as a result of its limited functionality in some areas (such as no support for any other attachment type other than static images).
    • The Lab is continuing to work with the Canny development team to try to improve things.
  • As was noted in the previous CCUG meeting, the placing of incomplete features (such as the glTF terrain and mirrors functions mentioned above) behind debug settings is a part of a more general move to add features and capabilities to the viewer more iteratively, allowing individual capabilities to be added (and potentially tested on a limited basis by those interested), but only fully exposed to all users once they have been validated / have any required supporting functionality / capabilities available to them.
  • There was a general discussion on the viewer UI and how it tries to meet the needs of very different groups of users  – content creators and “general” users, and how it might be made more responsive to specific modes of use – if you are using the viewer to explore, take photos, play a game in-world, and carry out basic building / editing, etc., then the viewer can be run in a “generic” mode to do all of these things, but if precision tools are required for design and creation, then the viewer can be switched to a mode which provides more specialised options through floaters and menus.
  • The question was asked that if SL is to support the “entire” glTF specification, including animations, does this mean future support for custom skeletons, and if so what does that mean for the current animation marketplace (e.g. preventing confusion as to what will / will not run against different skeleton types).
    • Runitai Linden’s viewer was that if custom skeletons are to be supported, then it would require suitable tools which allow for animation / joint retargeting (e.g. animations made for the current Bento skeleton could be retargeted of any other given skeleton, and vice versa). However, whilst there are tools to achieve this, they would require investigation and assessment.
    • There are limitations on how far retargeting will go, however (and it will not be entirely devoid of creator intervention), so if custom skeletons were to be allowed in SL, it was noted that expectations could be in part managed by specifying with which actual skeleton(s) the offered animation / avatar, etc. has been designed to work.
  • Concern was raised as to the constraints which might be applied if SL has an open gateway to importing glTF meshes from external sources such as SketchFab, where the content may well look pretty but is hardly appropriate for real time game rendering, the view being the Land Impact alone is not going to be sufficient.
    • The view from the Lab is that there will be some form of weighting system at import, at least analogous to Land Impact, but better suited for the acceptance / rejection of content, and this will vary for different content types (e.g. glTF scenes might be assessed of number of nodes, how many bytes in the asset, how many instances of the asset, etc.).

Next Meeting

† The header images included in these summaries are not intended to represent anything discussed at the meetings; they are simply here to avoid a repeated image of a gathering of people every week. They are taken from my list of region visits, with a link to the post for those interested.