macOS Universal Binary Viewer has the SLVoice plugin compiled for Intel chips only

[canny]

I'm a very recent user of an Apple Silicon (M1) MacBook Pro which I got as a gift from a friend. Fortunately, Linden Lab was so kind as to provide the SL Viewer with a Universal Binary version.

Unfortunately, the SLVoice plugin that enables Vivox voice communications is compiled as an Intel-only application. Simply put, unless you happen to be in a WebRTC Voice simulator server, voice will not work.

Currently, I don't have Rosetta 2 installed (and don't intend to!), so, perhaps unlike the majority of Mac users with computers bought since ~2020 — who had no choice but to use Rosetta 2 to run the SL Viewer.

As such, once they installed the Universal Binary, even though the 'main' code was now being running directly, the SLVoice plugin was using the Intel emulator from Rosetta 2. People (as well as your own team!) just hadn't experienced any issue, so the SLPlugin's functionality was never checked for compatibility issues.

Note that currently Apple Silicon users have only two options, if they wish to continue to use voice services in SL:

1. Install Rosetta 2.

2. As an Estate Owner, file a ticket to request a move of the region servers to the so-called "Second Life Preflight" version of the simulator server, which supports WebRTC Voice.

https://secondlife.canny.io/admin/board/bug-reports/p/macos-universal-binary-viewer-has-the-slvoice-plugin-compiled-for-intel-chips-on

---

This repo is using Opire - what does it mean? 👇

💵 Everyone can add rewards for this issue commenting /reward 100 (replace 100 with the amount).
🕵️‍♂️ If someone starts working on this issue to earn the rewards, they can comment /try to let everyone know!
🙌 And when they open the PR, they can comment /claim #5119 either in the PR description or in a PR's comment.

🪙 Also, everyone can tip any user commenting /tip 20 @canny[bot] (replace 20 with the amount, and @canny[bot] with the user to tip).

📖 If you want to learn more, check out our documentation.

[canny]

This issue has been linked to a Canny post: macOS Universal Binary Viewer has the SLVoice plugin compiled for Intel chips only 🎉

[GwynethLlewelyn]

I'd like to thank Dan Linden for his validation of this issue and for opening this ticket here.

As explained elsewhere, depending on how the build environment & framework of the SLVoice application works, this may be as simple as flipping a switch or adding a checkbox at some point in the compilation instructions/recipe/makefile/project. It's not as if it's something requiring a huge effort: it has zero impact on anything else and doesn't require a complex testing procedure: either it compiles and works, or it doesn't compile and the problem has no solution. It's as simple as that.

The main issue here is that SLVoice relies on Vivox libraries, which the 'public at large' doesn't have access to. According to Vivox, they have supported native Apple Silicon compilations for a year now or so (aye, it seems just to require a switch somewhere to recompile everything). But, of course, we don't have any simple way to figure out if there are any other complex dependencies which might prevent a compilation of SLVoice natively for Apple Silicon Macs.

The SLVoice binary that comes with every SL viewer (and I believe most TPVs as well) is almost two years old, thus predating Vivox's support for native compilation to the ARM architecture. It might mean that the libraries it's compiled with require replacement by newer ones, and this can have some consequences in terms of dependencies with other libraries.

Or, in other words: there might simply be too much effort involved in updating those libraries and making sure the compilation is robust across all platforms — focusing on deploying WebRTC Voice (which works flawlessly under the universal binary) might be a better strategy.

I have, obviously, a good, personal reason for originally making this request: I teach every first semester a class on virtual worlds, using SL as the platform. It was only last week that I realised that replacing my ancient 2014 Mac by a more modern version which already has an M1 Pro chip left me 'voiceless' — I had to resort to Discord to talk to my students. There are only a few classes left this semester, so, for me, either this is really just 'flipping a switch' and releasing a small patched version — meaning an ultra-short development cycle — or, well, for all purposes, I can only hope that WebRTC Voice is deployed grid-wide by September 2026. By then, all the above becomes essentially moot.

Thanks nevertheless for looking into this; I really appreciate it!