2018 SL project updates 4/2: TPV Developer meeting

La Vie; Inara Pey, January 2018, on FlickrLa Vieblog post

The following notes are taken from the TPV Developer meeting held on Friday, January 26th 2018. The video of that meeting is embedded at the end of this update, my thanks as always to North for recording and providing it. Time stamps in the text below will open the video in a new tab at the relevant point of discussion.

Viewer Pipeline

[0:00-1:25] The Voice RC viewer updated to version 5.1.1.512121 on January 26th, bringing it to parity with the Alex Ivy release viewer. The reaming viewers in LL’s pipelines remain as:

  • Current Release version 5.1.0.511732, dated January 9th, promoted January 17th. Formerly the Alex Ivy Maintenance RC – NEW.
  • Release channel cohorts:
    • Nalewka Maintenance viewer version 5.1.1.511871, January 17th.
  • Project viewers:
  • Linux Spur viewer, version 5.0.9.329906, dated November 17th, 2017 and promoted to release status 29 November – offered pending a Linux version of the Alex Ivy viewer code.
  • Obsolete platform viewer version 3.7.28.300847, May 8th, 2015 – provided for users on Windows XP and OS X versions below 10.7.

Forthcoming Updates

  • The Voice RC is the next in line for promotion to release status, and this could happen in week #5 (commencing Monday, January 29th, 2018).
  • The Animesh and Project Render project viewers both have new updates in QA.
  • There is a new project viewer for media handling (with updates to the latest version of the Chrome Embedded Framework) also due to be made available.
  • A further 64-bit update is in the works.

Environment Enhancement Project (EEP)

Project Summary

A set of environmental enhancements, including:

  • The ability to define the environment (sky, sun, moon, clouds, water settings) at the parcel level.
  • New environment asset types (Sky, Water, Days – the latter comprising multiple Sky and Water) that can be stored in inventory and traded through the Marketplace / exchanged with others.
  • Scripted, experience-based environment functions, an extended day cycle and extended environmental parameters. This work involves both a viewer updates (with a project viewer coming soon) and server-side updates.

Current Status

[12:25-15:43] Rider Linden now has a test simulator on Aditi where he is able to successfully apply windlights at the parcel level. The work will now allow windlight settings to be applied by height above ground, but this will likely be fixed ranges defined on the simulator rather than allowing completely arbitrary heights for different windlight settings, as can be done viewer-side with some TPVs at present.

General Discussions

Level of Detail / RenderVolumeLODFactor

[4:17-5:21] The latest Firestorm release, 5.0.11.53634, has altered the behaviour of the debug RenderVolumeLODFactor so that changes with set it to any value higher than 4.00 will not persist across log-ins (see my Firestorm release overview here for more). The Lab is now considering implementing a similar change within the official viewer.

Land Impact / Avatar Complexity Calculations

[5:33-5:52] I’ve referred to this project a number of times, which is intended to gather more representative data on the actual cost of rendering object in-world and the cost of rendering avatars and their attachments, with a view to revisiting the formulas used in calculating them, to see if anything can be done to make the calculations more representative of the “real” rendering costs (allowing for issues such as the potential to break content). It is currently hoped that the Lab will have enough data on this in the next for weeks to start carrying these investigations forward.

[27:13-28:12] One of the reasons often cited for creating one very detailed level of detail (LOD) model and a very low LOD model, rather than a number of LOD models (high, medium, low and lowest), is that the latter can penalise the finished model’s overall Land Impact (LI). This is something the Lab is going to be looking at in order to try to remove / reduce disincentives to making optimised content for SL as a part of this work.

[28:35-30:25] These updates are likely to affect the LI of existing and rezzed content. However, they will not be made in Q1 of 2018. Instead, the Lab plan to carry out simulator-side testing to ensure whatever changes that are decided upon cause as little disruption as possible, and will offer a strategy to help people transition to the new system to try to avoid any unnecessary item returns, show the revised formula run the risk of some content being returned when implemented (this might even comprise an increase in land capacity, if the difference between the “old” and “new” LI calculations aren’t too big). Improved tools for understanding LI are also planned as part of this work.

[31:08-31:47] To help people understand LI, content building, etc., Oz Linden has been trying to get more of the Lab’s own documentation on design and building to where it can be seen in public. This also extends to making more of the Lab’s own scripts being made available for public viewing.

Camera Presets Project

[7:09-8:00] Jonathan Yap is moving ahead with his camera presets project (see here and STORM-2145). This will most likely include Penny Patton’s recommendations for improved camera placement as one of the default sets of camera presets.

AIS Project

[8:51-10:28] The Lab is initiating a further Advanced Inventory System (AIS) project. this project will initially be focused on bug fixes, then will include looking for opportunities to deprecate old UDP-based inventory operations in favour of AIS – this work will likely take several months to complete, once started. Once any patches related to this have been made available to TPVs (with time given for them to be implemented), inventory UDP messaging will be turned off at the simulator end.

Texture Rendering and Caching

[16:05-17:15] Kitty Barnett has been looking at texture fetching/decoding and has noted some issues around discard levels in the viewer. Oz Linden would like to learn more on this.

In the meantime, the Lab’s own attempt to re-work how texturing caching works in the viewer hasn’t produced the results LL had hoped (e.g. improving the amount of textures which can be loaded per second). However, tests will continue in the hope that improvements can be gained. If the work is successful, then the Lab will look towards improving object caching as well.

Inventory Folders and Load Times

[18:52-20:50] Inventory and folders:  the recent change (current SL maintenance RC viewer and Firestorm 5.0.11) to include folders in the total inventory count has raised concerns about increasing inventory load times.

The Lab’s rule-of-thumb remains one of balance:

  • Individual folders with tens of thousands of item in them (object, links, sub-folders) – including trash – can cause the inventory load process to freeze, due to the way folders are individually loaded at log-in. Therefore, these are best avoided.
  • Conversely, having an individual folder for every single item – or just very small number of items can cause lots of little fetches that don’t achieve very much.

Therefore, somewhere between these two extremes is preferable.

Other Items

  • A resource has been made available to update the wiki documentation on the new viewer log-in screen widget.
  • There is liable to be something of a focus on the render pipe (there’s already the rendering project viewer in the pipeline), as a resource with SL rendering expertise has returned to the fold from Sansar.

Next TPVD Meeting

[1:35-2:05] The next Third-Party Developer meeting is set for Friday, February 16th, 2018 (although in theory this should be a date for the Web User Group meeting, which usually alternates with the TPVD meeting).

Advertisements

One thought on “2018 SL project updates 4/2: TPV Developer meeting

  1. I saw recently an object likes a Radiator , 1 li … it smells exactly the prim reduction by playing with lod … and Oh! surprise, it was done by the blogger Beq … seriously , before gives lesson, learn first ! or explain why at lod high the number of poly is upper 6000 and at average level it is at 56 and the shape of the radiator is totally broken Learn first, blog after !

    Like

Have any thoughts?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.