Ubisoft's Outland / Sebastian 'Class of 2009' Schauman

Sebastian Schauman graduated from CG Arts in 2009. Since then, he's been working at Housemarque Games. Two years of hard work have come to fruition recently with the release of their games title, Outland. I asked Sebastian to provide the CGAA community with an insight into his experience of readying Outland for release - and being the nice guy that he is, he has this to share with us:

"So Outland is finally out. My last 2 years of professional work can be summed up into a 7 hour play-through available to all for about 10 bucks a pop...

It's funny really to actually see it out in the open. It's nowhere near the game we initially started out creating and if I'd had to sum up the last 2 years in a single word it would have to be "iterations". The development process turned out to be a lot of trial of error and re-design over a long period of time. It's really not a time-efficient way of doing things, but as this was such a unique game-design, with a genre with a huge and critical fan-base, it required a lot of iterations to get right.

Outland was created as a smaller side-project, with the majority of the studio working on Dead Nation during the most part of the development. Our team consisted out of about 6-13 guys -depending on the timeline. Our core-team consisted of one lead game-designer, three programmers, two level designers, two concept/2D artists and two 3D/Maya-artists: a tiny work-force by today’s game-industry standards.

I joined the project straight after graduating from CG Arts. At that point they had already finished the initial prototyping and even presented Ubisoft with a presentation that had peaked their interest. The next step was to produce a FPP (First Playable Prototype) and get Ubi to greenlight the production. My first day at work I got a look at the silhouette 2D metroid-style shooter that it was back then and was put straight to work modeling 3D spiders based on earlier "2½D" spiders used for the prototype. I was initially hired purely as a modeler for the company, but pretty much as soon as I finished my first spider it became apparent I had to rig the enemies as well. Soon after, I had to start animating them as well and create all particle-effects for the game and help out "dressing" the levels - all of which I welcomed quite happily since it meant more experience.

As time went on the game underwent some quite dramatic changes; our hero's laser-gun got swapped for a sword, his magic powers swapped for Ikuraga-style alignments, single-player only got co-op additions, and so on. This was basically due to the amount of testing we did in the beginning. After trying certain things out it became apparent it would not make for a very unique experience or even a fun one, so we adjusted our direction. Art underwent iteration after iteration; although the main aesthetic (shadow-puppetry against vibrant colors) remained very similar to the initial concept, adjusting and tweaking of the visuals never really stopped until a few weeks before the game was launched.

As mentioned earlier, I did a lot of different tasks on the visual side for Outland. My biggest responsibility - and perhaps the task I found most satisfying - was enemy-creation. If anyone's interested in a typical pipeline for creating a new enemy, it went something like this;

• Our concept artist would come up with a basic concept for an enemy. This concept, along with others, would be presented to our group and the actual design of the enemy's behavior would be discussed and defined. After the enemy's design was at a level with which people where happy, the concept artist would do some further rework, tweak the image and send it over to me.

• Initially I was given fancy orthographic paintings from different views and so on, but as the dark shadow of our deadlines crept closer those paintings were reduced to a single image or even just a quick sketch. From there, I basically started blocking out the character doing some basic modeling. Once satisfied with the silhouette of the character, I'd sceenshot it and send it back to the concept-artist, who'd then do some paint-overs to further develop the silhouette. (Outland having the shadow-puppet visuals it does meant silhouettes being good was key). After that it was back to me for further modeling, UV mapping and texture-painting. All textures where then hand-drawn in Photoshop and obviously only utilized diffuse and alpha-maps due to the restrictions that outline-visuals have (i.e. no dynamic lighting, which means no normal, bump, or spec-maps).

• After that, we'd often do one more paint-over of the character and fix any remaining issues with the visuals. At this point, I usually started rigging the character, and depending on the enemy, made some quick place-holder animations. (i.e. idle-, walk-, attack animations). All this was then committed over to a programmer, who'd smack it all together into a prototype enemy viewable in-game. This followed a lot of tweaking to get the enemy behaving correctly, making proper animations for him, as well as adding additional behavior- and blend-animations. There was a lot of back and forth with the programmer to get code and art to work together. Once we both where happy with it, the character would be put into real game-play scenarios, after which designers would come to us with more changes and additions. Even after our in-house designers where happy with it, our producers at Ubisoft could even come back to us with additional changes they wanted made to the enemy - better visuals, animations, gameplay... In other words, there was a LOT of tweaking, both modelling, textures, animation and effects.

It's funny: the question my friends and family always ask is how can developing a game take such a long time - think about it - everything has to be made by someone. From every branch in the background, every sound a pebble makes and every particle that flies across the screen has to be made, placed, tested, and tweaked by someone.

I wasn't fully aware of the amount of work actually required to make a game when I first got into this industry. Game-art is not enough to get something looking good, it has to look good and work with the gameplay, code and whatnot. Just because you get something to work in Maya does not mean it will work in the game-engine. A lot of time is spent getting ideas working and coming up with ways to fake certain things in-game with the often limited tools at your disposal. Then there is the amount of time that is consumed by the never ending tweaking and polishing...

If it was up to the developers, games would probably never be considered completely done. (Unless perhaps if you're working for Blizzard and time and money isn't a concern). I can't play Outland without noticing hundreds of tiny graphic-bugs and things in need of fixing. But still, the fact of having a game out that you've had a hand in and hearing other players' reaction to it - especially the positive reviews Outland has received - is VERY satisfying!"

Congratulations, Sebastian! It's no exaggeration when I say we're all rather proud of you and we look forward to hearing about what you're up to next.

For earlier info dedicated to Sebastian's graduate experiences go here.