We started the New Year quickly, as we have a milestone set for the the start of February and we can't afford to lose any time working towards it.

Game development has always been a chaotic process. In the beginning you've had three blokes in a room, all of whom doing a bit of the art, a little bit of programming and some of the design. They got away with it because they sat next to each other and communication wasn't a problem. The game was only ever going to be played by gaming enthusiasts, none of whom were going to mind if the engine has been written by an artist, or if the sound was nothing more than a series of minimalist bleeps.

As teams and budgets have got larger, the culture has struggled to keep up. Games are immensely complicated and require strict, unglamourous project management. It's not just a question of efficiency - it's also about preventing bloodshed. Every games team has three very different and mutually antagonistic components in design, programming and art. Artists need the designers to have designed the game and the programmers to have finished programming before they can start their work. Designers would ideally like to know what technology they're going to have before they start designing their game. Programmers need to know exactly what the designers want before they start programming. All of which is a recipe for handbags at ten paces. And crap games.

But what do you do? You could always sacrifice creativity on the altar of MS Project but that wouldn't be much good either. You're creating a spreadsheet package - you need to leave room for the inevitable changes in the design. We've got a solution at Elixir called the Spiral cycle. Director Dave Silver explains: " The way we try to address this problem at Elixir is by using a spiral lifecycle for our development. Doing this involves breaking the development cycle down into stages of increasing size and level of completeness. At the end of each phase in the spiral there is an opportunity to adjust the direction of the project. We assess the major remaining risks, and plan out the next phase of development accordingly. This way we can have a full plan for each stage that takes account of all the work that needs to be done."

Compromise is also important. It's too easy for artists and programmers to become insular and parochial in their approach to work. The attitude becomes "well I've done my bit, it's their fault it isn't done." I loathe this sort of attitude - if you fail, you fail collectively. If you can encourage each side to see each other as allies then you've got a better chance of producing a good game and a happy team.

Over the last few months we've made a lot of progess with the art in Republic: The Revolution. Every couple of months we have an art review in which we throw away a lot of good work, in order to maintain consistency in the game world. In terms of the overall style, we've come a long way from where we were at the start. Initially many of the artists were dubious as to the fun factor associated with creating authentic looking Eastern European buildings for two years. We had a long discussion which is still going on. The point I tried to make is that reality isn't what we're looking for. Place like Kiev are in reality little more than adverts for foul Sixties prefabricated architecture. This is authentic, but is it cool? What I want are cities and towns that reflect the sinister reality of the game; imposing buildings that enforce the belief that the man on the street is a mere ant. We spent some time watching and discussing films where cityscapes have been changed or tweaked to achieve such an effect. Starting with Fritz Lang's seminal Metropolis, we looked at Blade Runner, Batman and Seven. The cities in each film are recognisably based on real places, but have been exaggerated to create an atmosphere conducive to the ideas within the stories. We also examined pictures of Prague and St. Petersburg. People have an idea of Eastern Europe that may not correspond to the reality - give them what they want. Even the rundown parts of Novistrana need to look cool in their own way. Miles of concrete monoliths aren't going to look good. We're trying to create a look that's plausible, cool and then realistic, in that order.

It's been a slow process and we've made mistakes (pink flats spring to mind….) but the art's getting better all the time. We discovered plans for monumental Communist buildings that were never built, such as Iofan's Palace of the Soviets (pictured below). We've placed them in the game to thereby fostering, I hope, the sinister film noir setting the game needs.

News of Republic: The Revolution seems to have ruffled a few feathers and several people have expressed doubts about Tim Clarke's Totality engine. In a creative industry there are many fragile egos and you run the risk of upsetting them if you attract what they deem to be unwarranted attention, all of which is very sad as one developer's success is rarely to the detriment of another. This has made us more determined to make the best game we possibly can. From bitter experience I've discovered that there's always a queue of people waiting to take a pop at you, hoping you'll fail. There's only one way to answer them - prove them wrong.

© 1998-2000 Elixir Studios Ltd. All rights reserved.