3Di Cloud starts at $10 per month

Japan’s3Di, Inc., best known for their browser-based OpenSim viewers, has set the starting price for their Unity-based, in-a-Web-browser virtual world at around $10 a month.

According to an announcement released yesterday, the “personal” 3Di Cloud starts at 9,800 yen a year, or about US $125, and can handle up to 500 simultaneous avatars.

An enterprise license starts at 29,800 yen for a three-month contract, or about US $380 — $127 a month.

A behind-the-firewall version of the software is also available, but pricing for this has not yet been determined.

There is no limit to land area or number of objects that can be placed in the world, 3Di senior manager Norman Lin told Hypergrid Business.

“Land area and objects are basically unlimited, since they run on the client,” he explained. “Larger worlds will require longer download times, of course. Content creators should tune their worlds to have appropriate size for their target audience’s hardware.”

The same principle applies to in-world scripts, he added. “However, if a script causes network traffic, that will cause increased load on the server.”

So, for example, a script that allows users to move in-world objects, or ride in vehicles, would require that the changes to the environment be instantly transmitted to all other users in the environment.

“Our tests with a fairly heavy-traffic test client indicate that we can support at least 500 simultaneous connected users on one server,” Lin said.

3Di Cloud’s default environment, now with a wider selection of starting avatars.

Price comparison

For individual users, 3Di Cloud compares well to other browser-based enterprise-oriented virtual world platforms, which typically start at around $50 a month.

The easiest of those is VenueGen, which offers a choice of pre-made virtual environments. The most developed platform is AvayaLive Engage — which is also backed by the biggest vendor, Avaya, Inc. — and allows for full customization of the environment.

Other browser-based virtual environments include the Unity-based Jibe from ReactionGrid, Second Places’ Unifier, and HostaVirtualEvent.com. Altadyn’s 3DXplorer is based on Java.

However, individual users may also be interested in Cloud Party, a browser-based social world where a private island starts at $15 a month and no plugin in or download is necessary because it runs on WebGL and HTML 5 Web standards.

The Unity-based worlds all require that users download a Unity plugin, and the Java-based world requires that Java be installed.

In addition, there are two Web-accessible OpenSim options.

SpotON3D has a proprietary plugin that allows a full OpenSim viewer to run inside a browser window. New users still have to download the full OpenSim viewer — in addition to the plugin — but the way it’s embedded in a Web page may provide some psychological comfort to new users. Low-prim regions start at $26 a month. Prices top out at $60 a month for a full 15,000-prim region.

Kitely allows customers to post a link on their websites that takes them to their Kitely world page. There, new users are prompted to install the Kitely plugin as well as an OpenSim viewer if they don’t have one yet. The plugin then automatically launches the viewer and sends the users to the customers’ region. The viewer does not run in a browser window, however. All users get one free region and two free hours of use a month (six hours the first month), with paid plans topping out at $35 a month for 20 regions and unlimited use.

Feature comparison

3Di Cloud currently has less functionality than any of the alternatives listed above, though users can add the functionality themselves.

“As it’s built on top of Unity, you can have the 3D models, animations and scripts that Unity supports,” said Lin.

For example, customers can edit terrains with the Unity terrain editor and import mesh objects and mesh avatars saved in the FBX or Collada formats.

No in-world building or scripting tools are available, however — everything takes place externally, using 3D design software, or the 3Di Cloud Developers Kit.

Say, for example, you wanted to provide vehicles for your avatars. You would need to create appropriate animations for the avatars, and import the 3D mesh model for the vehicle. Then you would need to write the vehicle script, using either JavaScript or C#.

“You would need to write a controlling script to move the vehicle and to force the avatar position to follow the vehicle,” Lin said. “We don’t have any sample code for this right now, but it’s definitely possible.”

Similarly, 3Di Cloud avatars cannot move objects around. However, customers can build this functionality themselves. This would require creating a user interface inside Unity that allows users to move objects, a script that sends out a global message that the object has been moved, and another script that listens for those messages and moves the objects for all connected clients.

“You would also need to define some policies and mechanisms for resolving editing conflicts,” Lin added.

Other features not supported by 3Di Cloud include avatar inventories, friend lists, offline messages, profiles, groups, voice, or in-world media. Customers that need this functionality would have to set up external databases to store this kind of information, and pass it into 3Di Cloud environments by scripting.

“We’re doing exactly this for a project in development right now,” Lin said. “3Di Cloud takes care of real-time in-world synchronization, and the Unity-based client uses information from both 3Di Cloud and also from an external information provider.”

But pulling Web pages onto in-world surfaces would be more of a challenge, he added.

“It’s currently difficult to do this properly and completely inside Unity,” he said. “It’s probably best to embed Unity within a Web page, and cause Unity to open a browser popup window that then visits Web pages in a synchronized fashion, transmitting the current URL to all other connected clients.”

Unlike previous 3Di products, which were marketed primarily to Japanese customers, 3Di is looking at a global audience for its 3Di Cloud product, Lin said, with documentation and support available in English as well as Japanese.

For companies that plan to be doing a lot of custom development, and expecting a large number of simultaneous users, 3Di Cloud may be a good fit. For companies looking for an off-the-shelf solution with a larger set of existing features, VenueGen, AvayaLive and Jibe would be better bets.

maria@hypergridbusiness.com'

Maria Korolov

Maria Korolov is editor and publisher of Hypergrid Business. She has been a journalist for more than twenty years and has worked for the Chicago Tribune, Reuters, and Computerworld and has reported from over a dozen countries, including Russia and China.

  • > “Land area and objects are basically unlimited, since they run on the client”

    I think this is very significant. The server-focused model of Second Life and OpenSim just makes things very expensive – after all, practically all our users have machines that can run physics on them (if they can run graphics, which they must!).

    We would need different technology though. Unity is nice but it costs money and is closed source. We do have open source projects that run full 3D environments on the web, like BananaBread,

    https://developer.mozilla.org/demos/detail/bananabread

    (which I have worked on). This could be tailored to run a virtual world, and would do the heavy work entirely clientside and use only free and open tools (zlib, as free as is possible). And it wouldn’t need plugins, instead it uses WebGL and JavaScript.

    • @kripken:disqus
      I looked at the code and it dose appear to be on the track we would want to go for getting Opensim regions in a browser without a plugin. The shooter app you have been working on looks impressive – nice work!. I should like to know more about tailoring it to run a virtual world. Many of us that run Opensim would like a browser application that doesn’t need a back-end viewer download and webGL dose look like a promising solution. Thank you for the link and bringing it up here. I do want to know more and I think some funding might be available too if what was said about crowd funding previously is anything to go by.

      • This approach isn’t OpenSim compatible I am afraid – it has a different structure for the contents of regions (no prims – instead it uses an octree of modified cubes), different networking protocol, different scripting, etc. It also has different content generation tools, but while different they are similar to Second Life’s in that they work in-world and not separately (press ‘e’ in the demo to enter edit mode, where you can modify things).

        In theory someone could write a conversion tool that takes an OpenSim region and converts it into a format this technology can run. But that would be difficult and probably not result in the most efficient data, since each world format has different things that work well in it – SL prims won’t be optimal in the cubes used in this engine (and vice versa, of course – this engine can do things prims don’t do well).

        So I guess this is only interesting for people that are willing to use a different technology than OpenSim. The benefits are that far less resources are needed on the server, that the viewer and not just the server are permissively licensed open source, and that the viewer runs either natively (might make sense for running on an iPad) or on the web in HTML5. So I think it has a lot of advantages and that’s why I’ve worked on it, but of course I understand if people that are used to OpenSim would not be interested.