Day: February 11, 2026

2026 week #7: SUG Leviathan Hour

Posted on by Inara Pey
Jade Koltai: Inis Oírr, January 2026 – blog post

The following notes were taken from the Tuesday, February 10th, 2026 Simulator User Group (SUG) off-week meeting (the “SUG Leviathan Hour”). These notes form a summary of the items discussed, and are not intended to be a full transcript. They were taken from my chat log of the meeting, and Pantera’s video is embedded at the end of this article – my thanks to her, as always, for recording and providing it.

Meeting Overview

  • The Simulator User Group (also referred to by its older name of Server User Group) exists to provide an opportunity for discussion about simulator technology, bugs, and feature ideas is held every other Tuesday at 12:00 noon, SLT (holidays, etc., allowing), per the Second Life Public Calendar.
  • The “SUG Leviathan Hour” meetings are held on the Tuesdays which do not have a formal SUG meeting, and are chaired by Leviathan Linden. They are more brainstorming / general discussion sessions.
  • Meetings are held in text in-world, at this location.

Mesh Face Count Mismatch Bug

  • This is a long-term bug whereby some mesh objects fail to accept a texture change on some faces. The texture change shows up on the viewer, but if you take to inventory or relog the texture change vanishes.
  • The problem appears to require changes on both on the simulator side and on the viewer.
  • Leviathan describes the bug a a form of protocol bug introduced by an old asset-upload-pipeline bug, which he believes has been complicated by a viewer-side bug.
  • In short:
    • LL pack texture data using a variable-length scheme, depending on whether all faces are the same or there are differences, but the actual number of faces is not encoded; instead the simulator has been relying the number of faces being implicit to the “VolumeParams” of a prim (this approach pre-dating mesh).
    • LL then introduced meshes with up to 8 texture faces.
    • This, combined with a potential bug in the asset upload pipeline which caused the number of faces in the asset to be wrong, caused errors on the server side of things.
    • A fix for the upload problem was implemented, but the the issue persists for mesh objects uploaded prior to any fix for uploads being applied.
  • Leviathan has a fix for the simulator issue(essentially a means to encode the number of faces into the legacy texture encoding without breaking protocol and without eating many bytes).
  • However, in testing it, he found a viewer-side issue in loading and updating / recording mesh face textures data. He’s still working on this issue, but believes the viewer and server will have to have an updated protocol interpretation so they can communicate how many faces they think there are in an object. the data will live in the new object created by making any changes to a rezzed item.
  • He noted that such a fix might also correct the issue of LSL not seeing the right number of faces.
  • This discussion also involved some confusion over sculpties and their  total number of faces, and the actual total number of faces (8 or 9 – it’s 8 hard-coded on the simulator).

In Brief

  • Leviathan has a fix for the problem whereby when sometimes rezzing an object on a mesh surface will fail and supply an incorrect or misleading message (e.g. not having parcel rez rights or something). See: FIRE-15429 and Beq Janus’ blog Coming to Firestorm soon… A couple of new features for builders and non-builders alike. It is currently in a Pull Request.
  • There are no updates on the status of the game_control work,  and the pre-release of the game-control viewer remains unchanged.
    • Moving this project forward is dependent on the progress of the new SL Linux viewer build.
    • The next steps are to make some UI adjustments and identify any further changes required to get the code through QA for release.
    • As a reminder, game_control provides some keyboard mapping options from the existing avatar control actions:
      move forward, strafe, turn, jump, fly, and (maybe) crouch. It doesn’t currently support “all the keys” input.
    • The reason for game_control is to provide LSL scripts access to game controller input. But how the controller interacts with avatar-control, fly-cam, and keyboard mapping complicated things.
    • There was talk during the early days of the project the game_control could be made to handle MIDI inputs; however this is currently not supported.
    • This led to an extended discussion on key mapping options.
  • Leviathan expressed an interest in turning to support the current drivers for the Connexion 3D Mouse devices (e.g. SpaceNavigator, etc.), given the current working driver for SpaceNavigator is badly out-of-date, and more recent versions. However, this is an interest, not a commitment to actually do so.

Date of Next Meetings

  • Formal SUG meeting: Tuesday, February 17th, 2026.
  • Leviathan Linden: Tuesday, February 24th, 2026.

† 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 rooftop of people every week. They are taken from my list of region visits, with a link to the post for those interested.