'Machinima' gains acceptance from filmmakers

I don't think so, there is something about participating in a shared space that makes SL compelling when it comes to collaboration.

Also, the virtual universe is the ultimate idea space, where the mind and imagination reigns supreme.

I win in SL because I am right, not because I am white and look like the typical all american CEO on a video cam.

As for controlling each others computers.. I don't see that. If it's necessary (rarely is) I'm sure you can jump into VNC in another window.

What would be ideal, of course, is functionality which supports whiteboards and better textual support. HTML on a prim will help, but shared drawing on whiteboards would be very very cool.

I've tried to get a large-scale production going in SL...

believe me, while the ideas sound nice, the fact of the matter is -- there are still large boundaries to adoption.

Organizing volunteers for one, is next to impossible through SL -- without direct contact, people have a hard time taking it seriously. Afterall, they log out and SL essentially gets "turned off."

Next, there are precious few people in SL with the hardware and willingness to act as cameras. This is why I've been suggesting "puppet cameras" for quite some time now. It would be nice to script a camera to do a nice sweeping pan while I do something else with my AV. However, from a technical POV -- I can see that being next to impossible in SL. At least with current bandwidth limitations; unless the developers can find a way to statically cache areas on a client, puppet cams would basically open up 2 streams to a single client... 1 stream already has enough trouble fitting through the pipe of a typical 3mps connection. Let alone rendering 2 streams on a single client machine!

Animating is another hurdle. In all my efforts to get "Hangin On," off the ground -- prematurely I imagine mind you -- the biggest difficulty after finding camera persons was animators.

Animating is not easy. It's by and large the most difficult thing to do well in SL. Anyone can pick it up.. but it takes a lot of time and is very tedious. To make an animation look believable without motioncapture is an art that few possess. And to create enough animation for even a short is a lot of work! Pierce is right when he says that until we can capture our ideas directly from our mind, we probably won't see large-scale adoption of SL for machinema.. unless of course we find ways to attract highly dedicated groups.

For now though, we may see small productions popping up now and again. I myself haven't entirely stopped. I just bought a new video card so that I can render SL at a decent framerate for video and have all the a/v editing tools I need. I hope to start a small production on my own and work up from there.

But if I were to try to get Hangin On going again, I would most definitely do it differently... and would start by pulling together a group of dedicated people I knew face to face and could meet up with from time to time. As idealistic as it sounds, SL really isn't a great place to collaborate with people from all over. We still have some hurdles to overcome in that area... and no amount of technology will solve that problem.

I completely disagree on the collaboration part.

I've worked on lots of stuff with tonnes of people and the collaboration has been amazing.

Hell, complete strangers will walk up to me and we'll start building something together.

You simply can not do that anywhere in any medium in existence other than SL.

In terms of the camera views:

Well, my idea is that you should be able to script something so you can simply play it over and over again, so you can capture with your main camera.

NPCs would go a long way to letting you script something over and over again.

But, yeah, it'd be nice if you could control it in a more smoother fashion.

As for the animations, yes, that's very true. However.. Where else can you even effect the animations of the avatars?

Depends on the collaboration -- yeah it's easy to just start building.

To put a film together requires a lot of work both in-world and out... and unless a person has a strong relationship to you -- chances are they're not going to feel the least bit obligated to the project.

Now try and fit all the different skills you'd need to make a decent machinema production in SL -- animators, props/costumes, sets, camera, actors/puppets, post-production.. not to mention the cost of sharing/hosting all that video during the post-production stage.

Now try collaborating as easily as it is to fly up and press a couple buttons to rez a cube...




NPC's.. I dunno. I hardly see using NPC's being any different than using other AVs as puppets. One would have to animate both only there'd be less scripting when using an AV since a live human on the other end can take care of a lot of the tasks that would take code a fair while to do.

Cameras? Yeah.. not likely that we'll ever get more than the one we all have. That's our one stream... anything more than that might not be technologically possible at thsi point.

And yes.. animation are tedious and I still think it's quite amazing that we can make our own in SL. Does that dimish the cost of producing those animations in terms of time and skill? Not at all.

My interest in NPCs is to make the world more fulfilling, not to replace real participants; I agree completely with other posters that if there are people interested in playing the parts, they should. I think it would be a really interesting thing to develop NPCs (even bots that could not obviously be distinguished from real players) and it is something I would like to work on, but that's another thread

Real actors voice animations at present - CG or even clay! - so even the lip synch isn't a problem; just tell them the original dialogue was SLenese or something. Only for real closeups could it be a problem, if you are a perfectionist.

I think the real crunch is control of the camera. I would love LL to release an SDK or even the source to the client that would allow a cut down object to allow both NPC development: having an A.I. Server interact with the world through a streamlined client; no rendering required, direct access to system messages and chat for natural language parsing, etc. However, the real advantage of such access would be to have a client that could be used to stream the view of the camera and allow movement like that available through the current player interface:

for example, cursor key control of the camera would allow pan zoom and dolly shots. Following a particular actor (as is evidenced in pressing the tab key during a chat, which allows you to track someone). The really exciting thing for me would be the ability to video stream the output back to SL, so a telepresence would be possble. Fly your camera object while seated in front of a screen. Pre-define paths according to specific locations, orientations and timings. and so on. This would really allow the creative use of SL to be realised.