I think there's no one more apart from Project Wonderland. I talked with them some time ago.
Remember that their license is GPL with classpath exception. I did not understand why don't they just use LGPL, so beware of possible legal pitfalls when modifying it for commercial projects.
If you just use Wonderland as it is, without modifying the core, it's perfect for commercial projects.
And also if yours is a non-profit project.
Pretty much anything that is built with a Java base should be as well, which would include but not be limited to...3dXplorer, and the Project Wonderland mentioned by Hyperia3d. Vastpark is intended to be cross platform by all accounts, though I'm not positive if they have extended their reach that far yet.
I'll dig through my files and see if I can come up with any more, but those I am pretty certain of, off the top of my head.
(I should mention that the above listing don't really take into account things that may simply have different versions of their program available for different operating systems and therefore be usable by different platforms.)
I'm becoming more and more interested in Blender. It can be used to develop 3D models and their animations, but it also has a standalone part (blenderplayer) that can play those worlds without the large, main application. Blender interfaces directly with python (one of the easiest to use and fastest growing languages today) so that worlds can be scripted to do arbitrary things, including networking.
The down side is that there seems to be a fairly steep learning curve. It may be possible to use a simpler approach to learning to use Blender than my approach, but I don't yet know what that would be. After I become proficient in using it I'll try to provide a way for others to get easier access to its capabilities.
A few things that make Blender worth investigating:
- it is free as in "free beer" and as in "freedom" (open source, LGPL)
- uses physics (arbitrary collision, hard and soft body, cloth mechanics, simulated forces such as wind)
- no need to buy a separate modeller application
- skeletal animation of avatars using forward and inverse kinematics
- deforming avatar skins
- it is fast
- cross platform (MSWindows, Linux, Mac, FreeBSD, Solaris)
- able to use all the libraries that come with python
- can import (and export) a wide variety of existing model formats
- has a very active and dynamic community
There are some things to beware of though. It has been developed for more than simply realtime 3D. A lot of what Blender does is meant for slow, photorealistic rendering of movies. That's great, but it means that if you want to do realtime 3D you need to be aware that many of the tools in Blender are of limited use, and some are of no immediate use at all. This contributes to the complexity of the program and can be frustrating when it is found that a pretty model looks different when run realtime. But this also has an unexpected benefit. For example, light glare is mostly intended for slow, offline rendering, however some modern, high-end display cards now support it in realtime. So designing for offline rendering can actually benefit realtime 3D because they are designing for the future.
Thanks for that assessment of Blender Miriam. I had not thought of Blender player for a cross platform VW technology. Python scripting and the many platform implementations are a plus. No server or explicit VW components would be cons.
It seems to be in a similar category then to OGRE and its derivatives, although OGRE is a rendering API not a tool. Also OGRE is LGPL and if Blender is GPL as Hiperia says then I would bat for OGRE as an API. There are a wide range of projects/code bases to use with OGRE too, games, python, etc. Not sure if there are VW APIs on OGRE.
Yes, I have been testing Xj3D on and off over the years but it has not got to the stage I would be able to use it. It is great that it is LGPL and Alan, Justin and others have done a great job developing this Browser.
However, Xj3D and X3D have not met my needs over the years and the attempts I have made to get some MU VW direction into X3D have not progressed well enough. I tried, for example, thru chairing the Network Working Group, to transform X3D from having portable networked content support to having strong portable networked content support, but this effort has kind of slowed to a stall.
Needs change too. I raised the idea of a smaller language for the mobile platform, derived mostly from X3D (or as a profile of it) and with perhaps some Collada features. OpenGL went through a kind of catharsis with the development on ES and ES2 for mobiles. These are like profiles of OpenGL and a lot of dead wood was removed, architecture changes made and some new stuff added to bring it up to date with recent shader hardware. The result, in my opinion, is a much more powerful, light weight and relevant version of OpenGL. I think X3D should have gone thru the same process.
The portability that comes with java is very attractive, but I honestly have never found a java 3D application that *generally* performs near to, say, the BS Contact Web3D browser, with roughly similar CPU load.
I did not noticed a strong difference of fps between BS Contact and Xj3D, if you remove the FPS limiter that comes per default. Anyway, higher than 24 fps no human can notice the difference, but if you are really concerned with performance, maybe you'll need a C++ library.
May I ask what for do you like it? I mean, do you want a library (like Xj3D) for developing a solution from scratch? Or a more elaborated system like Darkstar that has already most of the classes prepared? Or an already done solution that you just need to build worlds and use it?
If you specify your needs, we could recommend something better. There are many things out there.
yes fps comparisons can be a bit difficult to nail down and it is not worth discussing without a more direct scientific comparison.
as to what I would like it for:
I am interested in MU virtual worlds so the client /server would have to support that basically. Something like wonderland / darkstar seems to fit the bill tho the test I recently did with wonderland - it just crashed and I have to dig into the documentation to work out what is going on. At first glance there did not seem to be anything describing whether you had to setup a database first, with accounts etc so I just got a login window and a lot of class exceptions in the client console window.
As to what I need:
Preferably client and server designed for VWs, but at a minimum a client renderer and basic network support for client/server client/client comms.
An open source licence like LGPL so I can modify partrs of the software as needed.
I am interested in the merging of social nets with VWs so the other thing I am looking for is a way to integrate the graohical VW with a SN system.
Mutliple platform support. The first thing I get asked with out VW based on BS Contact is : "Can it run on a Mac?", which, of course it cannot.
FRIDAY NIGHT TALKSHOW Presents:A special concert/interview with singer:NESTO SILVERFALL discussing his involvement with music performance & his debut album.[9:30pm SLT/PDT] [LIVE/VOICE]http://slurl.com/secondlife/Cevedale/70/22/2538
FRIDAY NIGHT TALKSHOW Presents:
A special concert/interview with singer:
NESTO SILVERFALL discussing his involvement with music performance & his debut album.
IM Wurlitzer Seisenbacher for info.
[9:30pm SLT/PDT] [LIVE/VOICE]
http://slurl.com/secondl…
What Can the AVW Do For You?
Please send your thoughts and ideas directly to either Dave or Edita.
Virtual Leadership
The Association of Virtual Worlds provides leadership through an advisory board, member committees, publishing activities, educational programs, special projects, and other collaborative activities. To learn more and get involved click here.