There is a very little known feature of Second Life to do with showing video in world, this is the ability to show different videos to individual avatars on the same land parcel. I’ve mentioned this to a quite a few people now and all have been in disbelief, even some of the Lindens seem to be unaware of this feature.

The method to do this however is not a hack and has in fact been documented in the LSL API for as long as I know.

The method in question can be discovered by looking at the documentation for llParcelMediaCommandList if you have a close look at the parameters there is one called

PARCEL_MEDIA_COMMAND_AGENT

the description for which is “Applies the media command to the specified agent only.”

So lets look at a quick simple example.

We need a screen that can listen for urls over chat and then set the url for the person speaking. ( The screen must be owned by someone who has media permission on the land )

The script for the screen is …

integer listen_handle;

default
{
state_entry()
{
listen_handle = llListen(10, “”, “”, “”);
}

touch_start(integer total_number)
{
llSay(0, “talk on channel 10 to set your personal video for this land”);
}

listen( integer channel, string name, key id, string message )
{
llSay(0, “Setting Video play back to ” + message);
llParcelMediaCommandList( [
PARCEL_MEDIA_COMMAND_URL, message,
PARCEL_MEDIA_COMMAND_AGENT, id,
PARCEL_MEDIA_COMMAND_TEXTURE, (key) llGetTexture(0) ] );
}

}

Here’s a quick side by side screenshot of two AVs watching a different movie at the same time on the same parcel of land.

Extra points for guessing what movie trailers they are watching

Lots of interesting news coming out of the Virtual Worlds 2007 San Jose conference…

First up it’s the announcement of a joint partnership between IBM and Linden Labs to work on VW interoperability standards and technology. The official press release can be found here.

  “As the 3D Internet becomes more integrated with the current Web, we see users demanding more from these environments and desiring virtual worlds that are fit for business,” said Colin Parris, vice president, Digital Convergence, IBM. “IBM and Linden Lab’s working together can help accelerate the use and further development of common standards and tools that will contribute to this new environment.”

The second piece of news that caught my attention was the upcoming release of a custom SecondLife client from The Electric Sheep Company. Who have tied  in the release of their viewer to a cross-platform initiative around CSI: New York.

Usability of the current SL client has long been an issue and is one of the factors blamed by many for the high attrition rate of new users to SL.

The SL client has actually been open sourced for a number of months now, so it might be surprising to some that the release of an alternative or modified client has taken so long. However when more closely examined the reason is fairly simple, the open source client base has been a moving target for the OS development community with Linden making changes to the underlying network protocol used. This has now changed and the protocol template is now static, hopefully allowing alternate client developments to proceed with more confidence. Thats not to say the OS community haven’t made some fantastic contributions to the client, mainly in the areas of bug fixing and stability.

Now that the SL protocols are more stable and OS efforts such as the client, libsl and opensim are gathering momentum we’ll start to see some real progress towards a more cohesive story for VWs. The main thing now is to try and reduce the barriers to entry, and that means easy installs, better UIs and better integration with web browsers; so who’s going to be first to create a VW client running in a browser plugin ? Metaplace seem to be taking up that mantle currently.

If you’re aiming for a more professionally produced machinima piece then the following tips are for you, in no particular order…

Hardware
Use a beefy desktop if possible. You need a decent graphics card  and plenty of RAM. A fast hard disk really helps capture smooth video, slow disks and low RAM will cause you headaches as your buffers will fill up before the video can get writen to disk.

Record in as high quality as your machine can manage
If you’re filming on a decent desktop machine and it can cope then go for 1024p resolution, this will give you the best quality playback if you’re projecting onto a large cinema screen or using a High Definition plasma/LCD display. Go for 720p as the next best option (this is what i use most of the time)

Playback at an Event (trade show/conference)
Always playback from a computer hooked up to the monitor/projector. dont be tempted to put the machinima on tape or DVD the resolution is much worse than you’d think. If doing a live broadcast event using the video have a machine for backup, as last resort use a tape or DVD backup.

Hiding the UI
Unless you’re doing documentary or event capture where you need avatars chat text you’ll want to hide the UI, do this using CTRL-ALT-1

Avatar chat
If you’re filming a scripted piece then put the captions on afterwards, the standard SL chat is small and hard to read when playing back at a lower resolution save your viewers eye strain and add the captions afterwards along with any other titling. This also gives you the option to quickly do changes when your client comes back to you and requests a script change (which they will).

Things to tell your client
Make sure they know that filming machinima is a vary similar process to RL filming, you’ll need props, actors and a script. You’ll have to do  multiple takes and post editing. It all takes time. If you are able to have influence on the script etc try and keep the number of characters in each scene to a minimum.

Actors
As mentioned before, if you can keep actor numbers low do so, if you’re lucky enough to have access to multiple machines use alts and control the characters yourself when numbers permit. I use three machines on occasion which lets me use 3 avatars with one being the camera.
If you do end up using a lot of other avatars try and get them to all go on a conference call. its much easier directing by voice rather than typing when you’re trying to use you avatar as a camera.

SecondLife tools
Filming Path HUD is essential for getting smooth camera shots - panning, top shots etc. it’s the best tool around and allows you to lay out the path you want your camera to follow. You can set a focus target on another avatar to smoothly follow them while they and the camera moves. You can also set static focus points so that you your camera will stay fixed looking at a set point even when moving through complex curved paths etc.

Be creative with angles (but don’t over do it
Filming machinima allows you the freedom to use creative angles more easily than RL filming. try shots from above, below, panning, swooping and aerial. make sure  they will fit together nicely when you edit afterwards though i.e. keep continuity in mind.

Don’t forget composition
If you’re new to filming or photography then look up some of the composition rules of thumb, leading lines, rule of thirds etc.. once you know these simple guides you’ll spot them turning up again and again in hollywood films. They’re not hard fast rules but they work as a starting point and will help make your machinima a more visually pleasing experience for the viewer.

Film more than you need
play with different angles and capture scenes more than once. You can edit afterwards and pick the best pieces to use, if you have a script change then some of the excess video you’ve captured might come in useful.

Hope some of these come in handy in some way, if you have any more tips then please add them in the comments

Well I haven’t done any machinima posts of late, but I can feel a few in the offing having just completed one concept video for a client and with the promise of a few more lurking not too far in the distant future.

The machinima I’ve been doing has generally been of the type you use when you can’t do a live demonstration of SecondLife, it’s been produced to either illustrate potential scenarios/concepts or show case existing business aspects of secondlife. The venue for displaying this machinima has been at trade shows or during large presentations.

Video and particularly Machinima seems to keep cropping up in my day to day job at the moment, a client I’m currently working for has a need for a streamed video service for use at events. We want something reliable but cheap that we can use for the occasional streaming of a live event etc. into Second Life (any suggestions welcome

The one service I have been playing with at the moment is a startup called Veodia (veodia.com) which is currently in beta. It’s very similar to another beta startup
called ustream (ustream.tv) the one distinct and important difference (with my SL oriented slant) is that it makes available the video stream in quicktime format. Meaning that it can be streamed straight into Second Life with no conversion faffing or trouble finding dedicated servers to run a Darwin streaming server on.

Ian pointed me at Veodia a few weeks ago, however it took my account registration email some time to materialise (it was trapped in Hotmails spam catcher) so I was a bit slow off the mark in checking out what format the streaming was done in. Kevin Aires as ever the video streaming investigator got in there and found it was quicktime compatible and blogged within IBM about getting a Veodia stream into SL.

After finally getting into my account yesterday I started up my softcam desktop streaming app, set up a quick feed on veodia, grabbed the rtsp link from a view source and set the media url for my land parcel in Ukanipo.

Here’s the result a slightly disturbing world within world effect…

.

So what you are seeing here is the result of softcam capturing my desktop video display output, streaming it up to veodia and then it being streamed into SecondLife. Infinite loop complete. (ps the quality is better when you don’t go for this wheels within wheels effect)

So whats this useful for ?

-video conferencing: pipe your webcam into SL without the need to host your own streaming server.
- live demos:   do desktop sharing to a crowd in SL so you can do demonstrations from other apps (shared browsing or 3d app tutorials anyone?), simutaneously visible on the web stream too.
- world tours: get all your guests into one place with a screen in SL and then disapear off streaming your secondlife journey back to the assembled SL crowd.
- video link parts of SL: running a live event over several islands ? then provide video linkups between each of the islands so everyone can see what’s going on.

best of all veodia which is in beta is currently free, oh and it records too

Just stumbled across this in the sldev mailing list (ooh how retro ) from Argent Stonecutter, a cool looking open source avatar creation project called MakeHuman definitely looks worth following. The user interface seems particularly nice, I need to do some more investigation to see what kind of skin creation features it offers…

ooo check out the pretty sunsets coming our way soon in Second Life…

The interesting part is that Linden purchased Windmark, they must be feeling a bit more comfortable of late financially which is a good thing. I have to say purchasing existing technologies to improve SL is a good tactic, the existing Linden developers would be hard pressed to put in the time required to develop these features from scratch.

The ever inventive and productive Dave CJ from our emerging tech department at work has been hooking up more of his demonstration lab to SecondLife. Tracking the state of everyday objects and visualizing their state in world allows for remote control of devices and a quick way to check up status of a remote site/home or factory type installation. There’s a nice detailed write up over on UgoTrade..

Google recently acquired a company called Neven Vision whos description on Googles blog is as follows

Neven Vision comes to Google with deep technology and expertise around automatically extracting information from a photo. It could be as simple as detecting whether or not a photo contains a person, or, one day, as complex as recognizing people, places, and objects. This technology just may make it a lot easier for you to organize and find the photos you care about. We don’t have any specific features to show off today, but we’re looking forward to having more to share with you soon.

Neven Visions technology has also been used in conjuction with Logitech to produce Facial movement tracking technology for use with standard webcams, this sort of stuff would help really bring avatar to avatar interactions to life.

This stuff might look like kids play, but there is some serious comp/sci theory going on in the background here.

Another clever play in the virtual worlds space for Google, or just a coincidence ?

OK I’m pulling this straight from 3pointD, but this could be a biggee. Sony showed this video at the Game Developers Conference, a PS3 based vitual world using the latest physics and graphics engines.

In comparison to SecondLife it looks stunning and a lot more technology rich, the user created content aspects may not be there but the social aspects certainly seem to be. The stated ability to meet friends and head off into multiplayer games is a powerful one indeed, as a result the environment will no doubt gain traction and may become a strong differentiating factor for the PS3.

Anyway take a look at the video and form your own opinion.

A nice little article over at FoxNews on the official response from the Linden Labs lawyer to the blogger Darren Barefoot. Darren put up the Second Life parody website Get a First Life and received an official Un-Cease and desist reply from Linden. Good to see Linden labs leading the way in legal policy regarding humour