Last year I turned to the iPhone platform. It seemed an excellent starting point for an aspiring developer: make almost anything and distribute it worldwide for just 100$. Ogre 3D also had a very promising port in development, which was great since I had already spent a couple of years working with it. So I began to create a very basic prototype back in November.

8 months into development, and what started out as something simple has become an ambitious project. I never planned anything, so it has pretty much evolved over time. To be honest, I never thought it was even possible on the iPhone.

I took a break early June, and set myself the task of releasing something -anything- into the app store as fast as possible, provided it does what it says on the tin, with no bugs or crashes. 2 weeks later I had done just that. On the 14th day I submitted it to Apple, and by the 20th day it was in the app store. It was a great learning exercise about the process of creating an app from start to finish, using purely the iPhone SDK and Objective-C. Most importantly I got a taster of what the submission process is like.

It’s called Urchin – the app store link is here. It might keep the kids quiet for a little while ;)

Since then I’ve returned to developing my ambitious 3D app. I’m working on it full time around the clock, making exciting progress. Be sure to follow me on my personal Twitter if you’re interested in the development side. Any announcements will be on TidalWare’s twitter.

I will be revealing some details about the project in my next post. Stay tuned!

What a difference!

Okay, new dev diary for you. If you’ve been following us on Twitter, you’ll know that we’ve still been very busy with stuff, including, yep, Project Utopia.

By the way, we setup a Flickr page for Project Utopia if you didn’t already know.

Those screenshots look neat, huh? However, we hit a bit of a dilemma. The terrain system featured in those screenshots looks impressive, but there’s various underlying problems. Not only that, but Ogre, the 3D engine we use, has a brand new terrain system in the works. Here’s a lowdown of the advantages/disadvantages:

Current terrain:
Advantages:
- Beautiful, powerful atmospheric system.
- Feature rich including paging, splatting, normal and parallax mapping etc.
Disadvantages
- The sky with atmospheric scattering is intensive and requires a decent graphics card.
- No dynamic weather.
- Uses the old Ogre terrain system.
- Shaders are commercial.
- Support is not great.
- Slow development.
- There are still many unsolved bugs.

Ogre’s terrain:
Advantages:
- Very efficient.
- It’s just a world object, not a “scene manager”, so that allows for a huge amount of flexibility, even with the possibility of seamless indoor/outdoor support.
- Great support and rapid development.
- Alternative sky has dynamic weather.
Disadvantages:
- No atmospheric scattering, though it will support it eventually.
- Other disadvantages are down to the fact that it’s still WIP.

So what do we do? Well, basically, it “makes sense” to go with the Ogre terrain, because it’s going to be the best choice in the long term. But that means we have a lot of work to do to catch up with the graphical quality of the old environment.

Here’s where we’re at so far:

Ran the Visual Leak Detector, and these were the results:

There aren’t literally that many memory leaks, it’s misleading – simply failing to call delete can cause dozens of reported leaks.

Background loading is where resources are loaded seamlessly as you play. The client connects to a HTTP server where the resources are located and downloads them to an offline cache. When you exit, the next time you play the client will use the resources from the cache unless they have changed on the server. Most MMOs use patching to update game resources – this means long waiting times before the game even loads. We will still need some form of patching for binary files and major updates, but it will take considerably less time.

We have put our game engine and distributed server architecture aside for now and have decided instead to create a standard Client-Server system. This consists of a Login Server and a World Server. Each World Server will hold an instance of the game and support a maximum of 2000 players. This is a far more realistic approach for the short term, and will give us more time to focus on gameplay and art.

Thanks to Ogre 3D and a new landscape system the graphics are looking better than ever. All it needs now is some good 3D art, made easy with our editor which I will discuss in another post.

For your viewing pleasure: