Firestorm: where next and early looks

Update February 25th: As per a comment from Jessica Lyon, Firestorm have now merged the Server-side Baking code and updates to RLVa into one of their private repositories.

Update, 19th February: A transcript of the core part of the meeting, including Jessica’s Firestorm status overview and the Q&A session, complete with the video recording, is now available.

Update, 14th February: The initial video recording of the meeting is now available on YouTube, and an HD version will be available soon.

firestorm-logo

Wednesday February 13th saw the Firestorm team host an open meeting to discuss what is happening vis-à-vis Firestorm given all the various ongoing viewer-related projects currently underway (CHUI, materials processing, server-side baking, plus Firestorm’s own updates and improvements) – and when the next release is liable to hit the public at large,

Several members of both the Firestorm development and support teams were on-hand to field questions, with Project Lead Jessica Lyon leading things off with a 15-20 minute overview as to what is happening, where the viewer stands at this point in time, what the plans are for the immediate future and what we might expect to see in Firestorm in upcoming releases.

The Short Version

  • The Good:
    • Firestorm will be supporting all of the new viewer capabilities coming out of LL, although CHUI will require careful consideration as to what is adopted and how, as Firestorm already offers several similar options to those being added to the viewer by CHUI
    • Firestorm will be getting a range of new features (although not all at once) which include: further work on re-implementing legacy search capabilities, the ability to save and reload personal settings; more OpenSim support; new windlight settings; new UI skinning; further work on adding v1-style functionality
  • The Not-so-good:
    • Serious crash and other issues have also come to light in merging Firestorm with the latest LL 3.4.5 code which the team are endeavouring to resolve
    • Server-side baking (SSB) is the priority for the Firestorm team at present (as it is with other TPVs), as it has a major impact on how people will see things in-world, and it is the project which LL are emphasising. However, integration of the SSB code into TPVs (particularly those supporting RLVa) is not proving easy
    • The emphasis on work at the moment is overcoming bugs, issues and problems and trying to get Firestorm to a point where it is running the SSB code.

Taken together, the latter points mean that while a new version of Firestorm is in development, there will be something a further wait before it appears, and when it does, it my not have such a huge range of new features as has been found in previous releases and might suffer from stability issues.

Jessica Lyon
Jessica Lyon (seated centre, at the edge of the stage) with members of the Firestorm development and support teams, discusses Firestorm on Wednesday February 13th

Viewer Status

There are some serious issues within the Firestorm development code which are delaying progress towards a potential release. Firestorm has been merged-up to the Linden Lab 3.4.5 viewer code, and this has given rise to some severe problems for Firestorm (and is actually having an impact on other projects, as I reported earlier this week).

Commenting on the situation, Jessica Lyon pulled no punches, stating:

I’m going to be completely honest with you guys. Right now Firestorm, for us internally, is in pretty bad shape since our merge with Linden Lab’s TIP (3.4.5 code). There are a lot of bugs that we’ve inherited; there’s a lot of regressions which we’ve inherited. Ed [Merryman, lead for Firestorm Support] is crashing about two times a day – and for those of you know Ed, know that Ed never crashes. So if Ed is crashing on our recent builds, we’ve got some problems. We’ve got some log-out crashes, log-out things; log-in crashes … Basically, we’re not in great shape, and we’ve got a lot of fixing-up to do before we’re ready for a release.

As well as inheriting bugs, the merge has also highlighted bugs and issues within the Firestorm code itself which also need to be fixed. All of this adds up to recent builds for the viewer being “way worse” than the current release version in terms of stability and issues, and it is going to be a while before these issues are fully resolved.

Server-side Baking

Server-side baking is perhaps the most prominent viewer project underway at the moment, inasmuch as it is essential that all viewers connecting to Second Life be able to support it in order to avoid in-world experiences from being broken. Simply put, avatar skins and system clothing will not render on viewers which do not support SSB once the code is fully deployed, as shown below. )Things are somewhat more involved than that, and for those unfamiliar with the project, I’ve covered it in-depth in Avatar Baking: “and the clock has started!”. )

The SSB problem in part: I'm stabding on an SSB-enabled region. On the left  - as I appear to others who are using an SSB-enabled viewer; On the right, as I appear to others who are using a viewer which does not support SSB.
The SSB problem in part: I’m standing on an SSB-enabled region. On the left – as I appear to others who are using an SSB-enabled viewer; On the right, as I appear to others who are using a viewer which does not support SSB.

As it stands, Firestorm has yet to be merged with the Lab’s supplied server-side baking code for the viewer, although work has been underway within the team in a separate repository to the 3.4.5 code merge. A major problem here, as I again reported earlier this week, is that SSB has considerable (and negative) impact with RLVa. These problems are compounded by the fact that the test regions for SSB functionality are all on Aditi, which has considerable issues of its own at the moment, which are affecting people’s ability to reliably test code, and all have scripts disabled – which makes testing RLVa fixes alongside SSB somewhat difficult.

Currently, the Lab remains sympathetic to the issues TPVs are facing (and have offered help wherever practicable), and are not currently pushing a date by which TPVs must be ready for SSB to go live. They’ve also acknowledged that some of the problems TPVs are facing are down to delays on the Lab’s part, such as not making any bug fixes to the viewer code available until January 30th, some seven weeks into the planned eight-week window in which it had been hoped TPVs would be able to integrate the code. However, it is clear that TPVs are feeling under pressure to get SSB-capable versions of their viewers sooner rather than later.

Please use the page numbers below to continue reading this article