ambershee Blog

0 comments by ambershee on May 1st, 2008 digg this super bookmark

You may not have heard much out of me for the past few weeks - but that certainly doesn't mean anything is going on. For the past five weeks, me and some of the more veteran Shee Labs crew have been working our socks off to create a prototype game using Unreal Tournament 3. You'll probably never get to see it, but it will be showcased at The University of Applied Sciences and Arts of Zurich next week.

So what's the point? Well, it was an important exercise in team work (and a way of helping out a friend finish up their university coursework!) before we embarked on something a little bigger.

The next step, in case you missed the little recruitment drive in the forums, is a proper, public project.

It started out as a game mode, derived from the five week project, however it rapidly mutated into a monster. We had originally set our deadline to the third week in June - it all seemed so easy for our couple of programmers and our artist. Then design kicked in. We haven't even started yet, and we've decided to bring in the specialists, have doubled (looking to quadruple) the size of the team.

We're working on a small total conversion now instead; this isn't just a game mode for UT3, this is a whole new setting and way to play. It needs a complete scrubbing of the original assets, and the creation of entirely new ones. We're working on some new nuances in game dynamics, giving the visuals a much stronger tying in to the action; rather than having them a pretty overlay to the game - using advanced shader and camera techniques to signify more than eye candy. We even have a budget put aside for extra assets, are working on online ranking stables, and we're going to be filming live action sequences to overlay into the game environment! - something unheard of in a mod yet, as far as we're aware.

We're still aiming to release third week of June.

It's going to be a long stretch, but consider this the first (very quiet) official announcement of Shee Labs Game Development's 'DataRun'.

Until next time!

AmberZ

3 comments by ambershee on Mar 17th, 2008 digg this super bookmark

It's been a month now since I embarked upon the Mutator Week project. The premise was relatively simple; collect a group of people together who want to get to grips with writing code for Unreal Tournament 3, and assign a small project every two weeks and see what happens.

Initially, things look incredibly promising, with nearly twenty signups for the first week. I was hoping the theme (Light) wouldn't prove too difficult. Alas, despite the volume of signups, the number of active participants was very small, and we only got five mutators written on that first week - three of which I wrote myself.

The second week launched (Surprise), and things looked equally dim - I cranked out two mutators, and almost wrote a third, and there were three other mutators written. An improvement, no less, but perhaps not what I was originally hoping for.

Hoewever, the project is by no means a failure, and it certainly isn't going to be dropped any time soon. Working with those who are willing to contribute to the community project like they are is an absolute pleasure, and I'll continue to find the time for them. Hopefully, our activity will spark some interest within other budding programmers too. We're now entering our third week (Upgrade), and our weeks are now lasting longer (14 days, in fact). I'm hoping this will relieve some of the pressure  - a mutator in a week is hard work! - and give some extra quality to the project.

Now it's on to launching Mutator Week's much requested sister project - 'Map Month'.

4 comments by ambershee on Nov 20th, 2007 digg this super bookmark

Well, these rantings certainly haven't been rantings yet. Although I'd expect that to change in future...

Most notable, I put out a quick tutorial on getting started writing mutators for Unreal Tournament 3 today, prior to the release of the game itself. Hopefully that'll help some people get started with what is essentially a very rewarding aspect of modding.

Personally, I find that mutators can be some of the most rewarding things to program. They're very short term projects, with a much smaller workload than any pretty much any other kind of mod - a few simple lines of code, and maybe a little bit of extra media, and you're set.

Mutators don't require an enormous amount of effort, and can be a great testing bed for trying out new things without worrying about how much time you're going to lose to it. You can write mutators, and do some of the wierd and wonderful things you might have otherwise been afraid to try - although of course mutators do also have their limits.

It's writing tutorials for things like these that make you appreciate how useful such systems as the mutator are for learning experiences, something which Epic has certainly got right - and I'm sure we all love to try that new weapon replacemen or gameplay twist that someone has come up with.

There'll defiitely be plenty more tutorials in future - hopefully I'll be able to throw out at least one a week, and move into increasingly difficult ground, laying things down like the basics of creating new game modes, setting up frameworks for total conversions, and going through all those nitty-gritty sections that confuse everyone.

I'd love to set up a group of people dedicated to writing nothing but mutators, experienced and new alike. It's just the ticket, as people can learn from one another, work together to solve problems and of course, they're doing nothing but being creative.

Unfortunately, I get the feeling that the interest for such a thing just isn't there. People don't want to be writing the little mutators. These days it's all in the almighty total conversion; there's no glory for the small projects, and no appreciation for the and creativity behind them. New comers always seek to make the whole game, and never start small, something which is becoming increasingly impossible in this day and age.

So just how many fantastic small mutators are we going to see? 

9 comments by ambershee on Nov 13th, 2007 digg this super bookmark

Made some progress on what may be the first mod I release for Unreal Tournament 3 during the many long and horrible buildings of native classes (yuck!) at work.

The project is under the working title 'Duck and Cover', and is a new game mode. The core elements behind it are the cover system, that is reminiscent of current games such as Call of Duty 4 or Gears of War. It's slow progress, but I might even have it ready on UT3 release day if all goes well, but I doubt it.

So far I have created a new pawn class, game mode and all the rest, as well as some special volumes. The volumes can be placed in the map editor to define where the player is able to take cover. The volumes can then be configured to determine which ways the player is able to lean around corners or over the cover to take a peek or shoot at the enemy.

The player is notified when they enter the volume, and they can press a button bound to the take cover move - they are automatically locked into position and will automatically crouch if the cover is low. They can leave cover at any time using the same button.

One type of cover will completely lock the player movement - this is used for things like pillars, barrels and the ends of walls, where the player has no need to move or if they did move they'd break cover. The second is currently not totally functional, but I intend for it to allow the player to only move parrallel (backwards and forwards) behind the cover. When they reach the end of the cover they may lean around it, if possible, otherwise they can lean over it.

All in all, good progress.

I'm dreading getting that second type of cover to work properly. I'm dreading even more trying to get circular cover systems working!

And then getting AI...

0 comments by ambershee on Nov 9th, 2007 digg this super bookmark

Well, it looks like version 4 is upon us. It's curious to see how it's rapidly turning into a MySpace / Facebook style application. There's a lot going on and a lot of empty space so it's obviously got a long way to go - but I'll give it a chance.

I've recently been reverse engineering the shader system in the Battle for Middle-Earth 2. I successfully managed to discover that the shaders used in the game are written in what appears to be HLSL and are compiled with the Microsoft D3DX9 Shader Compiler. Of course which version is used is another question entirely, but I should hope that the differences between Direct-X SDK versions are fairly minimal in that regard.

Having discovered this, I found with great glee that the 3dsmax exporters included with the Command and Conquer 3 SDK allow us to export materials using complex shaders in the same format. Bonus!

All I need to do now is work out how to implement them in the game, seeing as they can be written, compiled and exported forit.

Toon shading anyone?