Youtube link

More to come in the near future.

How does this fit into the scheme of things you ask?

We have various technologies vying for market penetration of their respective 3D plugins, of which Unity is just one. Google themselves are developing an open source standard for this medium and, until recently, seemed to have an advantage in that it would surely be supported natively by both their Chrome OS and internet browser at some point. The recent development, however, compromises their opportunity to monopolise this sector but, in the scheme of things, promises for a more inclusive browsing experience outside of their dedicated open source 3D content delivery system.

To quote Unity developer Jonas Echterhoff...

Yes. A very big step actually. In fact, the Lego Star Wars game Google is showcasing at the native client booth at the I/O conference right now has been tested on Linux (Ubuntu), and worked just fine! 

It does, of course, require running a browser which runs native client (that is, Chrome, for the time being).

So there you have it, official word on the subject. This gives Unity yet another edge against competing open source standards which are likely to be implemented with Mozilla Firefox. As ever, the future is uncertain but the war is happening now and will dictate the development environments of the future.

I leave you with an official Google video on the subject.

Until Unity allows more than diffuse maps for the terrain shader, it will always be difficult to make good looking snow. Specular/bump maps would have allowed for glistening in the sun and made it appear less flat. I tried to combat this via the use of particle emitters for snow being blown downhill (Modern Warfare 2 uses the same technique), 'volumetric' light and fog. Summer uses an atmospheric engine and not a skybox, it's an altered version of a script I found on the Unity forums last year (more on this in a later post). The grass is factored into the real-time AO and, while it incurs a massive performance hit, looks great.

I might write a tutorial on this in the near future as it's something a lot of people have been asking me for.

Interestingly, the lighting seemed to look better without the roof, and so that's how I decided to bake it. You might notice that there's no discernible light source from which to cast the shadows.

It took me about a day to do. I'm quite pleased with my current workflow and hope to upload more content in the future, along with tutorials.

If anyone's interested, I lightmapped the scene with lightup and have written a tutorial on the process. You can read it here.

If you have the Unity 3D plugin installed, you can walk around the scene here.

Regarding the new GPU, there's not much sensible information out there currently other than that the chip itself seems to be a multi-purpose solution to the systems architecture. The main processor is an ARM Cortex-A9 and the GPU is an ARM Mali GPU. The CPU and GPU are all on the same chip, running at 1GHz in comparison to the iPhone 3GS which runs at 0.6GHz. Finding information on the GPU is tough, though the manufacturer's website states it supports 4x and 16x full scene anti-aliasing (should you need it).

Of importance to note here is that this is another opportunity to make some money. Creating apps for the iPhone, which sell, has become significantly more difficult due to how much content there already is in the store. While the iPad will supposedly support all the current iPhone apps, the new SDK ensures that there'll be loads of new stuff. The faster GPU also means that we will have fewer constraints to deal with when optimizing content for the device and more opportunities to impress users with visuals. The future is looking very promising with new markets and a growing industry, exciting times indeed.

I'm going to try and keep the polycount low on this one as it's going to be delivered via a web browser and you can never tell how fast another user's machine is. To that end, I've allowed the user to switch between quality settings and might make some super low poly models for the ultra low setting.

Again, I'm not using real-time lighting, as beautiful as the stuff is. This is mainly in consideration of performance issues. Having said that, there's a lot you can do with baking your own lightmaps, as you can see in the shots.

Just a quick reminder to readers. Although Unity 3D is marketed as being a game engine, it is just a 3D engine. You can do more with it, other than make games.

I recently came back from Rome where we (our group from the University of Hertfordshire) attended the first in a series of meetings related to the European AVATAR project. It aims to establish which conditions in a 3D virtual world are most conducive to the practice(s) of learning and teaching. To that end, one of the requirements is to establish which virtual world is best suited to the task. Readers of the blog will be aware of my work in Second Life and how, besides development,  I have been undertaking research into how the medium is able to facilitate communication and learning.

One thing on a lot of peoples minds is how we define a virtual world (for the purposes of education). Evidently, a simple 3D Virtual Learning Environment will not suffice. The task is to find (or event construct) a 3D multi-user environment with an integrated set of building tools and a freely accessible library of 3D models. This sounds very much like an exact description of Second Life, the only problem being that there is reluctance in adopting the platform across the board.

For all its problems, Second Life has shown us what happens when you give people, with no background in 3D, a simple construction set and the means to work together. Some of the things people have built range from entire virtual settlements to railway networks spreading over significant areas of the landmass. There is no longer any doubt that this medium has removed the most stubborn barriers (lingual, social and taboo alike) and encourages cooperation between individuals. The only problem lies in its implementation.

People do not like the idea of Linden Lab (the creators of SL) being in charge of all the content contained inside the environment. It's too big a responsibility for them to handle and, to be frank, they charge too much money. SL has gone out of its way to recreate a lot of the problems which exist in the real world, namely a lack of space. This is virtual space which is distributed at a cost, it's almost a license to print money (were it not for server maintenance fees it would be).

One of the most important requirements of the virtual world we need to find/create is that it be open source. Again, Second Life's open-source equivelant, OpenSim, seems to be a likely candidate. Time will tell what the final decision will be, but rest assured there's going to be an extremely critical analysis and that SL no longer has a free ticket.

I leave you with a video of some of my previous work in SL.