Developer Diary #10

[Mal]

ART

Wow what a month it's been we are now so close to the end of pre production you can almost taste it.....MMmmm It's a bit nutty!

My last week or so has been involved with the User Interface (UI) and it's been an interesting journey reaching our final look and feel, which near as damn it is ready to be implemented.

So what's involved in creating a UI. Well it is one of the hardest design challenges in games, especially in RTS games where the player spends more time using the interface than other game genres. If the UI isn't well organised, frustration can set in and the enjoyment of the game is lost.

First of all the design is conceived on paper outlining the various panels and option open to the player during game play. Once these are identified we can proceed with a pre visualisation of how it will hang together and animate on screen. This Pre-Vis stage should be fast to produce and easily editable. You want changes to come thick and fast!  I decided to use XSI (our 3D package) to do the Pre-Vis in as I can work very fast with it and it has a built in compositor which allows me to build up elements in layers, giving me a lot of control, and the ability to make changes quickly. This means that the output will be linear and not interactive. This isn't an issue for us as the design on paper is in good shape and we already have some elements working in game.

An image of the Compositor in XSI.

Each 'Over' node represents a seperate UI element.

In XSI I arrange each individual element of the UI into a render layers, so I can treat them separately in the compositor later on (see above). I'm not concerned about the look of the UI at this stage so I just build it out of boxes, which is super quick. I then animate the UI simulating the players actions, opening the "mini-map" and "objectives" panel for example. Some of the work requires textures to show the contents of panels changing. For this I animate texture support objects over the boxes, essentially sliding the texture over the box to find the part of the image I want. When I'm happy I'll render it out into a quick time movie and present it to the team, we'll discuss changes and I'll produce the next iteration.

An example movie the team would see (Quick Time required)


Once the Pre-Vis is locked down we can move on to the visual aspect of the UI.
During this stage of development I like to act as an external design agency viewing the rest of team as my client. It's perhaps a strange way of thinking but it helps me to be objective and not be precious about the designs. Like in the Pre-Vis we employ a system of design, review and iteration.

The brief was to create an aesthetic for the UI which complements the game, reaffirming the Visual style and story elements. We have our Retro-Futuristic style to consider which is a great starting point, however finding a direct or obvious route to an elegant UI with that alone won't be easy. We needed to find another element with a strong design direction which can help things along. We introduced some Art Deco principles into the mix and that worked very well. Art Deco had a big influence in Early Sci-Fi, most notably in Metropolis (1927) which arguably kick started Sci-Fi as we know it today!



Above is an example of the design iterations we went though. Although it may appear that these took a lot of time, they were done over the course of one day. (Thank you - Photoshop layer styles for making life easy! ;-)

As I mentioned at the start we are almost ready to implement the UI. We'll show off the full designs later on when it's in and working and utterly ace!
Until next month then....

[Mike]

Hmm...Mal appears to have used my usual intro. Comments on how busy the month has been? Check. Promises that pre-production is nearly complete? Check. What does that leave for me? Um.....nice weather we've been having, eh?

He's not wrong though, it has been a busy month and we are so close to completing pre-production that...ahh, you know the drill.

So, what exactly has been happening with the design? We took a step back and looked at the game as a whole, to see if the way we had the game working at the moment was giving the sort of feel that we'd wanted right from the start. When you're working on a game for a long time, it's very easy to get tunnel vision: you spend so much time working on all the little details that you forget to make sure that they're all pulling towards making the game the way that you originally wanted it to be. In essence, that's one of the most important roles that a designer has to perform; keeping their "eyes on the prize" so to speak.

We looked at the speed of the game, not just how fast the men run or how quick the combat is, but how often, and in what manner, information is presented to the player. That can have a huge effect on how the game is perceived: if you present too much information too quickly, the player can easily be overwhelmed; present too little too slowly and the game can seem very lethargic and the player can feel detached from the action. Of course, some of the systems that we know are going to be feeding information to the player aren't in the game yet, so we had to fill in those gaps in our mind, but we started to get a pretty good idea of how the game was going to feel in play. You can get a sneaky peek of some of the systems we decided on if you look carefully at Mal's UI animation above.

The other main thing we looked at was the story and mission structure. It's too early in the production process to be locking down exactly what's going to happen in each of the missions, but it's very important to have an overview worked out. The reason is that our programming elves need to have an idea of the different types of objectives we want to include, so that they can build the systems into the game that would allow us to set those objectives up (it's an interesting Catch 22). Likewise, the art elves need to know what kind of settings we're going to want in each of our landscapes so that they can start working on concept art for them.

To this end, we came up with a list of cool places to set missions, and a very basic outline of the entire plot. I've been working to bring those two together and set out the mission structure of the game as a whole, along with the main objectives for each of the missions. The mission structure is another very important element. You can set the missions up in a strict linear fashion with the player encountering one after another; that type of structure allows you to have a very detailed plot as you always know the order that the player will experience the story elements, but it can feel like the player is just being dragged along, not ever seeming to influence the world by their actions. The other extreme is to have all the missions freely available at any time: this gives the player total freedom to experience the game in any way they wish, but obviously means that it's really hard to include an interesting plotline; after all, the player might experience the "end" before they've seen the "beginning".

Currently I'm working on a structure that's somewhere in the middle, with the player encountering small groups of missions that can be tackled in any order, with the groups being unlocked in a linear manner; hopefully that should provide the best of both worlds. Of course we may want to change this after we've seen how it plays, but it's a good starting point for now.

Man, I'm really jealous of Mal's flashy animations, but I don't think that a video of me typing in Word would look quite as cool. Maybe I should do next month's diary as a video diary?

*Millions of voices cry out in terror before being suddenly silenced*

No, okay, perhaps not.