
The following notes were taken from my audio recording + the video recording by Pantera (embedded at the end of this summary) of the Third-Party Developer meeting (TPVD) held on Friday, September 27th, 2024. My thanks to Pantera as always for providing it.
Meeting Purpose
- The TPV Developer meeting provides an opportunity for discussion about the development of, and features for, the Second Life viewer, and for Linden Lab viewer developers and third-party viewer (TPV) / open-source code contributors to discuss general viewer development. This meeting is held once a month on a Friday, at 13:00 SLT at the Hippotropolis Theatre.
- Dates and times are recorded in the SL Public Calendar, and they re conducted in a mix of Voice and text chat.
- The notes herein are a summary of topics discussed and are not intended to be a full transcript of the meeting.
Official Viewers Status
- Release viewer: version 7.1.10.10800445603, formerly the DeltaFPS RC (multiple performance fixes, etc), dated September 11th, promoted September 17th.
- Release Candidate(s):
- None at present.
Upcoming Viewers
- The next RC viewer due to appear is the ExtraFPS viewer. There are some bugs still to be resolved before this viewer can see the light of day on the Alternate Viewers page; like DeltaFPS, it is primarily focused on performance improvements.
- Further performance improvements will be made to the viewer following ExtraFPS, but work will also be transitioning to other viewer work. This next viewer after ExtraFPS will likely be the Maintenance B viewer. Details on it contents are still TBC, but it will hopefully include some of the work being put into getting Linux builds back up and running.
- After this, maintenance viewer updates will follow in the usual alphabetical patten (C, D, E, etc.), with viewer being given a suitable name in accordance with the current dinosaurs naming convention.
WebRTC
Summary
- The replacement of the Vivox Voice service and plug-in, with the WebRTC communications protocol (RTC=”real-time communication”). Roxie Linden is leading this work.
- Key benefits:
- WebRTC supports a wide range of real-time communications tools in common use (e.g. Google Meet), supporting audio, video and data communications, and is thus something of a “standard” approach.
- Offers a good range of features: automatic echo cancellation, better noise cancellation and automatic gain control, much improved audio sampling rates for improved audio quality.
- Opens the door to features and capabilities to voice services which could not be implemented whilst using Vivox.
Status
- Under the latest schedule, simulator deployment of WebRTC support is now set to commence during the week commencing Monday, October 7th. This means if all goes well, it could be fully deployed across the grid during the week commencing October 21st.
- The delay has been to allow LL to make further adjustments to the service.
- In the meantime, peer-to-peer and ad-hoc WebRTC can be tested on the WebRTC regions of WebRTC Voice 1, WebRTC Voice 2, WebRTC Voice 3 and WebRTC Voice 4. However, there is no bridging between WebRTC peer-to-peer / ad-hoc and Vivox.
- During the transitional period when there will be a split between regions running Vivox and region running WebRTC, so:
- General spatial audio should work through the viewer, regardless as to whether a region is running WebRTC or Vivox.
- However, peer-to-peer, conference calls and group chat sessions might be subject to various disruptions (e.g. if you attempt to make a peer-to-peer call from a WebRTC region to some on a Vivox region (or vice-versa), the call will not go through; if you commence a call with both parties on a WebRTC region and one subsequently moves to the Vivox region during the chat, the call will be dropped, etc.).
- [12:42-14:00 and 24:04-27:45 – chat only] It was noted that Voice roll-off on WebRTC regions does not appear to be as effective as with Vivox, with people much further away still being heard.
- This was apparently an intentional decision, but can be further adjusted.
- It was noted that perhaps the roll-off for WebRTC should be adjusted, as it was felt the distance over which voice conversations can now be heard could discourage active conversations in Voice regions, due to potential problems of conversations overlapping one another.
Additional Changes
- There have been issues with the overhead Voice visualiser (the white do that displays “sound waves” when someone is speaking) always being red on viewers with WebRTC, and in muting and unmuting – both of these issues should be fixed in the upcoming ExtraFPS viewer.
- ExtraFPS also introduces a new Preference option for users to turn off the overhead Voice visualiser in their view if they wish.
- The switch is located within Preference → Sound & Media.
- The inference is that the option will likely be OFF by default (i.e. no visualiser shown), as the introduction of the toggle option is to “reduce screen clutter” – although this may be revised, subject to feedback.
- A further inference is that having the option off will not alter the Voice visualisers in the Conversation / People floaters.
In Brief
Please refer to the video below.
- [Video: 9:31-10:25] Graphics work:
- The focus remains on performance fixes.
- As per my previous TPVD Meeting summary and my week #38 CCUG summary, Geenz Linden is working with Rye Cogtail on the visual improvements (improved tone mapping, alpha blending, etc.) , with work also continuing on the anti-aliasing improvements.
- [Video: 10:26-11:39] The above led to a reminder updates to improve linear alpha blending for objects using Blinn-Phong materials; again, please refer to my week #38 CCUG summary for specifics.
- This work is being targeted at the Maint B viewer, and are said to be pending some simulator-side work (which itself should be in the Barbecue / BBQ simulator code update that will follow behind the WebRTC deployment.
- [Video: 16:51-18:33] A question was asked on the state of the pose tool developed by NiranV Dean.
- This was originally developed for Black Dragon and has been ported to Alchemy. It was also contributed to LL several years ago.
- It give the ability to pose an avatar (in your local view), such as for photography.
- The code has never been integrated into the SL viewer, and part of the reason seems to be the old chestnut of “shared experience”.
- However, LL are interested in developing it, and in receiving code contributions which may help overcome their reservations.
- [Video 29:14-32:46] There have been a number attachment issues making themselves felt (e.g. attachments vanishing on teleport arrival; attachments adding and then seeming to remove “after a few seconds”, etc.).
- LL are aware of a number of such issues and are investigating them.
- The fix for the attachments being lost following a TP will be deployed as a part of the WebRTC simulator update (which, as noted, is now due to start deployment during the week commencing Monday, October 7th).
- Those encountering attachment issues should check the feedback portal, and if they have a specific issue that does not appear to have been reported, file a report.
- [Video: 40:03-44:09 with extended chat discussion through to near the end of the meeting] Future viewer work:
- Brad Linden has been looking at native Apple / ARM support for the viewer, and considered this more of a “next year” project as the number of ARM users increases as opposed to those on Intel-based Apple systems.
- He also noted that he has been working on some of the 3rd party libraries required to build the viewer to make them more “universal” for use in different OS versions of the viewer. This should ease the viewer build process when working on multiple platforms, and he noted that contributions from TPVs who may have done similar would be welcome.
- It was reiterated that where Mac systems are concerned, LL is unlikely to implement Metal as n OpenGL replacement, but could look to a middleware layer to implement Metal compatibility for the viewer via Vulkan (or whatever is finally chosen to replace OpenGL for Windows).
- In terms of performance improvements beyond the current work, the Lab is looking to better profile hardware and then seek targets of opportunity where improvements might be made to general viewer performance.
- [Video: 50:03-50:28] RLVa contributions from Kitty Barnett: one pull request has been taken and is expected to make its way into the viewer Develop branch soon (if it is not there already), with more to follow.
- [Video: 50-29-End] A discussion on have the avatar selection process within the viewer point to the Senra avatars rather than the older “classic” basic avatars to help new users (there is currently a strong disconnect with the on-boarding process which utilises Senra with no real pointers for new users to understand the different avatar selection processes, Senra being entirely inventory-based).
Next Meeting
- Friday, October 25th, 2024 at the Hippotropolis Theatre.
† 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.