Remember that vividly realized alien playing Paganini I mentioned last week? Turns out there's much more backstory to that demo: The alien was actually created with a polygon mesh file made by a Resident developer who is part of Linden Lab's private Beta mesh program, working under a company non-disclosure agreement. However, last weekend there was a Linden announcement that the NDA had been lifted (at least temporarily), and now those private Beta developers are giving details:
- Import format will be Collada
- Collada allows rigging which means: Wearable clothing, entire avatar replacements that bend with animations
- Five submeshes for each object
- Four different Levels Of Detail
- One for collisions for Havok
Much more here. The biggest news is that the specific mesh file format is COLLADA, an industry standard that's compatible with Maya, 3D Studio Max, SketchUp, and Blender (among other major 3D creation platforms.) This will make Second Life content much more compatible with other 3D online worlds, games, and graphics (which may greatly encourage designers to try developing in SL.) COLLADA also means SL developers will have a means of legitimately backing up their content outside of Second Life, without running afoul of CopyBot controversies. For Second Life content consumers, this means more realism and detail to the content they enjoy -- including clothes, which will be easier to wear, and better fitting. As a source told me, COLLADA meshes will probably change Second Life content more than sculpties did.
When will SL developers get meshes to work with in Second Life proper? Last month, Linden Lab's former Chief Product Officer promised them this quarter. Would that were the case; they could not come soon enough.
Update, 12:30pm: I asked Linden Lab to confirm the COLLADA mesh details revealed by beta developers. And Linden PR rep Peter Gray replied: "We do not have any announcement or further info about mesh to share (or confirm) at this time."
Except the NDA being lifted was taken back and the people in the beta are back under NDA according to the full range of chatter.
But it is a nice tease to try to keep people interested in SL given the sudden interest in alternatives (inworldz grew quite a bit in the last couple of weeks, etc.)
I am curious how LL is going to deal with the fact that less prims will be needed for complex builds which means less land will be needed which means less tier for LL. Will LL change to charging by polygon used? And have a polygon pool for avatars to cripple all the jewelers and boot makers and in general make SL look as crappy as most of the "modern" video games?
Posted by: Ann Otoole | Tuesday, June 22, 2010 at 05:52 AM
Meshes have the potential to be a huge win for performance -- and SL's anemic graphics performance is a major obstacle to adoption.
But I doubt they'll have any impact at all on the avatar attachment market unless they look and move signficantly better than prim/sculptie/flexi constructs.
And if meshes free up prim count, builders will use more prim count. There's always something more that you wanted to do with a build that you just couldn't fit in the prim limit. Now you can, and builders will.
Posted by: Arcadia Codesmith | Tuesday, June 22, 2010 at 07:02 AM
Of course, the dark side of higher prim count is lower performance client side, and longer time downloading all the geometry to your client.
I don't know a lot of detail about the COLLADA format. It will be interesting to see what the limits are in terms of the number of polygons in one mesh. Right now, if you go to a very sculptie-rich area, it takes a long time for them to download (meaning you see lots of weird looking spheres until they do), and if your graphics card is a few years old, your frame rate goes to heck. The same thing may or may not happen with meshes, depending on how it's all implemented.
Posted by: Rob Knop | Tuesday, June 22, 2010 at 07:07 AM
Of course, the other thing to think about is -- if we get meshes, then everybody's going to have to move to viewer 2.0. I would be surprised if they implement the ability to look at meshes in the old viewer. Previously, it would have strained dev resources to support multiple versions of the viewer, and of course, now they have fewer dev resources than they used to....
Posted by: Rob Knop | Tuesday, June 22, 2010 at 07:08 AM
Another issue is the ability to learn and access the pro 3D tools. One of the major forces in SL is the ability to create without being Pixar.
*Consider an army of SL creators who can't afford the pro tools*
But we all want better graphics just calculate the price.
Posted by: Clausuriza | Tuesday, June 22, 2010 at 09:53 AM
I'm loving the potential for increased performance and better graphics, but dreading where most educators will end up. We generally lack pro-level tools or the skill to develop with them. That will vary school to school, of course.
Some of the teaching-and-technology folks who attend our roundtable could find a niche developing content for a new generation of SL builds and avatars. But will they have time and energy to do so? Or will educational builds that remain in SL be dowdy and pokey when compared to other content?
Now I can see one reason why LL has de-emphasized education. They may have reached the same conclusions I did and said "we cannot hold up progress for 8% of the users."
Posted by: Ignatius Onomatopoeia | Tuesday, June 22, 2010 at 10:24 AM
Well Bender is free and that is the main program I use for all my sculpty making. And COLLADA format is supported on Blender.
And Blender is quickly building its way up to being a strong application for 3d creation.
Posted by: Gattz Gilman | Tuesday, June 22, 2010 at 10:31 AM
Gattz, great point re: cost. But if even making textures in Photoshop lies beyond the reach of many faculty in higher ed, I don't know if they'll master Blender.
I might, but many of us don't tend to get much credit in our annual reviews for learning new software (or for using SL). I may be dead wrong about that as the reason why the US edu sector has not rushed in make simulations.
Posted by: Ignatius Onomatopoeia | Tuesday, June 22, 2010 at 10:48 AM
All this tech stuff simply amazes me...for gods sake...we can't get new users through that first couple of hours trying to learn how to use any viewer. Even the super trick, fantastic, do all, fix all, be all, disaster thats 2.0.
And now we want to add content creation tools that only tech savvy heavys can use?!
Stand by...Secondlife=Facebook Farmville on steroids.
Reminds me of any sleazy used car saleman dangling shiny keys in front of a gullible fool while steering him to the unrusted side of the junker.
Posted by: brinda allen | Tuesday, June 22, 2010 at 11:16 AM
It's guesswork at this point, but I'd expect meshes to suppliment prim building, not replace it.
And I'd expect to see a rapidly-booming market in "prefab" meshes, much like the existing market for sculpts, to serve the needs of those who don't have the time or expertise to roll their own.
Posted by: Arcadia Codesmith | Tuesday, June 22, 2010 at 11:24 AM
First of all, this is awesome, and long time coming. Second, the haters need to pull that smelly rod from you know where.
Linden Lab is ADDING new ways to support standard 3D format for content creation. Not taking away the in-world tools.
Cost is a non issue, Blender for 3D and GIMP for 2D graphics. Both completely free. They both have immense documentation on how to use them freely available on the internet.
The only people effected are those who do not have the time / skills required to compete in professional space.
Will be interesting to see what SL looks like in two years. When new content is created and spread through the grid.
Posted by: Not You | Tuesday, June 22, 2010 at 11:32 AM
First thing I thought of wasn't clothing or avatars.
I'm thinking tunnels, caves, mountains, snowdrifts, land bridges, and other formations that have better collision surfaces than sculpty versions do now.
-ls/cm
Posted by: Crap Mariner | Tuesday, June 22, 2010 at 11:37 AM
No one is going to take away you old tools for building prims, people will complain about everything, sheesh
Posted by: Metacam Oh | Tuesday, June 22, 2010 at 11:53 AM
All of these features sound great, I can't wait to give this a try. Surprised to hear about the multiple meshes for level of detail - that should make the pixel perfect builders happy.
It'll be interested to see how the balance between performance and detail plays out. Has LL has come up with some sculpty-style trick that will help all this new 3D info get to the viewer quickly?
Crap: yeah, a physical boundary that matches the sculpted shape will give us all sorts of new ways to build landscapes. should be fun :)
Posted by: Vlad Bjornson | Tuesday, June 22, 2010 at 02:05 PM
Very cool news!
Just one thing... this has nothing to do with Linden Labs, but rather the residents themselves. I always get annoyed when I see a sculpted prim being used to make a shape a normal prim could. I suppose it is inevitable that we'll be seeing the same with mesh imports. I just hope it won't get too out of hand.
Posted by: Fenix Eldritch | Tuesday, June 22, 2010 at 02:27 PM
I can't wait to see the CLOTHES!
Posted by: Angie Mornington | Tuesday, June 22, 2010 at 03:31 PM
Thanks for the link back to Slog, Hamlet.
http://secondslog.blogspot.com
Posted by: Frans Charming | Tuesday, June 22, 2010 at 04:31 PM
I've been waiting for meshes for some time now. This is actually very good news. If executed properly, it could be a real game changer, and that is something that is really needed right about now, more than ever.
Posted by: Little Lost Linden | Tuesday, June 22, 2010 at 05:42 PM
Somebody needs to learn how to keep their mouth shut.
Posted by: secret squirrel | Wednesday, June 23, 2010 at 08:01 AM
While SL's peers, such as WoW and Blue Mars, benefit from meshes, they prefer to distribute a mass of content on DVD and/or use one-time downloads of extra content when they're added.
While meshes are a good thing for SL, it's in conflict with two of its core features. The constantly changing landscape, and the the fact that when you visit one location you have to download all the content from its neighbours.
There might be a chance for some client-based solutions: longer downloading screens while content is actually downloaded - which most non-SL users would be comfortable with from other experiences; or being able to restrict downloading content to the parcel that you're visiting.
Having your avatar float in space, with blobs popping up before taking shape, and bouncing off things that haven't appeared yet, don't combine to create a good first-user experience.
Posted by: Lucius Nesterov | Friday, June 25, 2010 at 01:12 AM
Thats an interesting video but how was it made? I've heard that SL meshes will be static, but that video has an animated mesh with several spot lamps and multiple dynamic shadows. I really doubt it was made in a SL viewer.
Posted by: dahlia | Friday, June 25, 2010 at 10:56 AM
Kirsten's S20 viewer has a beta attempt at implementing realtime simplified multi-light shadows and Global illumination. Runitai has also been working hard on similar things in her time at the lab, and I would NOT put it past her to create the necessary cheats to allow this, dahlia.
Posted by: Patchouli Woollahra | Saturday, June 26, 2010 at 11:41 PM
sculpties and prims are very inadequate for real 3D artists who wants discrete controls of geometry and UV maps. This is why Games and things that generally looks better Dont use anything that Second Life uses. Which is nothing more than low level in game NURBS.
Posted by: Shinto | Saturday, October 16, 2010 at 07:39 AM
I recently came across your blog and have been reading along. I thought I would leave my first comment. I don't know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.
Posted by: Belstaff Leather Jackets | Thursday, January 12, 2012 at 05:43 AM