The Metaverse and 3D Web, as blogged by Mark Wallace and friends

“You can have the niftiest stage on the grid, but if your av is standing around looping the same twitch guitar animation for half an hour, you might as well just make an audio stream and be done with”

mmm…now that’a an interesting perspective on user expectations and what “live” music should be all about. I think it’s a bit sad if you judge the total worth of an SL live music performance on the constraints imposed on the performsrs by the current visual technology instead of the music. What about the interaction between player and audience that SL allows near real-time two way communication at a level beyond merely streaming to the web? [albiet still fairly narrow?] What about the spontenaity of a performer jamming something on the fly?

Granted a single loop animation can be dull to watch [and I load different ani loop sequences form HUD myself to try and vary this aspect http://thelaundras.com/SL/KomusoWeb/index.html, but I think there is more to a virtual performance than sitting on the floor looking up at the computer monitor/screen with a lighter in your hand staring misty eyed at your dream idols av jerking around at 1fps.

Maybe it also depends on the music style to a large extent, but from my limited experience people are chatting a lot [which is a big part of the social aspect of sl performances], moving their own av around dancing, or maybe just physically jumping around in rl or listening while doing something else and then running back to type something between songs…which is my preferred audience behaviour;-)
That said, I am working on enhancements to my own av hud to allow a little more randomness into the animation aspect while I play…within reason.

You also have to consider the performance limitations imposed by lag, sim av limits, frame rates…and what is feasible to do for a performer who is running sl, media stream encoding and upload, and audio software processing in real time on the same machine…it’s a cpu killer as well as a workflow issue to manage a live performance in sl…so spare a thought for us on the other end trying to do it instead of dismiss us so easily because we don’t emulate a rl performance down to the last nuance.

I’m sure duran duran will come up with some more visual element, as they are in the spectacle performance zone anyway and have the money to do the puppeteering angle [which I'm sure will happen] but for the solo musicians of sl it’s more about the music imo, not the puppet. Sometimes the whole movement that thinks sl must be an exact clone of rl makes me scratch my head…why does it have to be like rl?

I’m starting to wonder if you could be a successful sl musican if you had an ugly av…SL Idol anyone? aaaaaaaaaaaaaaaaaaaaaaaaaaarggggggggggggggggggggggggghhhhhhhhh;-)

Oh, man, Komuso… It wasn’t my intention at all to disparage or dismiss live performances in SL! If I came across that way, please chalk it up to verbal clumsiness, and not my true feelings on the matter. From an audio standpoint, the ability of a performer to interact with his audience in real time, via remote, is nothing short of amazing. It’s a huge step in virtual immersion, and I’m holding up my virtual lighter and screaming for more. And if that is done purely by audio, that’s cool. Perhaps I should have said “live, interactive audio stream.” The actual performances are great. I guess I’m just talking about the trappings.

[Brace yourself- the rambling now begins.]

My point was that if the performer is going to adopt a visual aspect for his performance – by placing his av on the stage and saying “this is me” – then there needs to be some thought and effort put into keeping this visual counterpart moving in an engaging fashion and (perhaps) interacting with the audience, as well. Otherwise, it can be a distraction and a detriment to the performance.

Even if you’re really grooving on the music, it’s jarring to look up and see the avatar standing there with her oddly-angled wrist bisecting the neck of her torso-eating guitar, looping through the same slightly spastic animation for the Nth time. We’re in SL, and we’re saddled with human-looking avatars standing on plausible looking stages. (For some values of “plausible.” Flat platforms with performers standing/sitting on them under the influence of gravity, anyway.) That comes with an expectation of RL-emulating behavior. This is true even if the audio component is the primary method of interaction with the audience. If someone can come up with a performance venue and an avatar representation that doesn’t intrinsically emulate RL stage performance in function (and good luck with that- I’m stumped), then perhaps this expectation can be removed. But, for now, we’re left with “eh, it’s SL” and a deliberate act of will to ignore the obvious flaws and limitations. That’s unfortunate.

Maybe it all boils down to “If you’re going to do it, do it well.” If you’re going to display your avatar as part of the act, then every reasonable effort (with “reasonable” defined by the limitations of your equipment and the situation at hand) needs to be made to integrate the avatar with the performance. Sounds like your HUD work is heading that way- sounds slick, anyway.

That being said, I kind of agree with you about the solo performer vs. the spectacle of an 80s hard rock band. Clearly, there is more need for visual performance alongside the music with the latter. It wouldn’t be Duran Duran without the jumping around and head thrashing (unless they’ve changed quite a bit since the last time I heard them). That’s just part of the whole performance, and it’d be fairly lame if it was replaced with stand-around-and-loop. I suppose two different types of musical performances (at least) are going to evolve in SL: the polished, scripted, pre-programmed audio-visual spectacles; and the freeform, less-visually-oriented, more intimate solo acts. There’s room for both. And probably the only real requirements for the latter are to plan ahead, practice the SL performance, make sure your (that’s a general “you,” not you specifically, Komuso) attachments fit properly, and perhaps change your posture to fit the songs. Even that minor effort would go a long way.

I guess if Duran Duran, or anyone else, wanted to throw money at the technical problems, they’d make two islands: an audience island and a stage island. That way, the framerate for the performer avatars/puppets would be tolerably high, and preprogrammed effects could be depended upon. I think that’d be the case- it’s hard to guess what the impact of 90 child agents (onlookers from an adjacent simulator) would have on the stage sim, but clearly it’d be less than the load of 90 avs residing on the same server.

Of course, if we can’t get more than 90 avs (or 100 or 80- whatever the limit is now) in one performance, this isn’t going to work for big names in the long-term, anyway. Maybe LL needs to come up with a performance grid, with a close-packed group of reduced area audience sims (say just big enough for 80 avatars to sit in a well designed seating system- 1024M^2 would be generous), adjoining a larger stage sim. Or, alternately, some kind of topological hocus-pocus adjoining multiple full-sized sims to the same border (the one between audience and stage sims). But that’s pie-in-the-sky. Perhaps that’s a better idea for a rental business when SL goes open source.

And as far as the ugly av is concerned, Frogg Marlowe seems to do pretty well for himself. :)