Shaping Ty the Tasmanian Tiger 2 for the Younger Market
January 19, 2005 Page 1 of 2
Ty the Tasmanian Tiger 2: Bush Rescue is the sequel to 2002's Ty the Tasmanian Tiger, which was also created by Krome Studios. For those of you not familiar with the Ty series or Krome, we are the largest Australian game developer, whose flagship IP is a character action platformer inspired by the thought-to-be-extinct Tasmanian Tiger. In developing the Ty series, we felt there was an opportunity to create a quality title - and character - that would appeal to the younger gaming audience while offering enough variety and interest for the masses. We also wanted to tap into and share with the world Australia's many offerings including a beautiful and diverse environment, unique culture as well as indigenous animals.
|Concept art of the Control Tower from Ty 2.|
As the producer on Ty 2, I oversaw everything involved in the game's creation, from design to level population. Looking at the sales and feedback from the first game, it became apparent that its relative success with younger gamers would affect the direction that the sequel would take - we had a fan base to please. Additionally, we had some big internal goals and expectations to meet.
It seemed the sequel to other platform games that were released around the same time as Ty 1 (Jak and Daxter, Ratchet and Clank) were being skewed towards a slightly older market, but we decided early on that we wanted to stay committed to the younger audience we had won over in the first game (the 6- to 12-year-old gamers). That said, we also decided that we needed to have the game "grow up" a little (as the players had grown up) with the introduction of additional age-appropriate concepts and features including more powerful enemies and level of difficulty. Everything in Ty 2 was made with the goal of being 'family friendly fun', so parents would feel comfortable buying this title for their children.
Aiming towards a young market with family friendly content doesn't mean you have to make a game without all the exciting features adult gamers come to expect in a premium title. This time around, we were determined to include some of the ideas that didn't make it into the first game (like more giant robots and vehicle based missions), and concentrate more on the "action" elements of platformer games. In addition to these enhancements, one of the other goals that emerged during the development of Ty 2 was the desire to change the game from having a stock-standard "hub with portals" structure to a more "sand-box" structure, where the player was able to do what they wanted, when they wanted. This style of gameplay seems to really appeal to the young gamer demographic with their shorter attention spans and love of exploration.
Another factor taken into consideration when we began development on Ty 2 was the forum posts made about the first game. While not the largest forum group around, the fans at Down Under are quite vocal, and we made a concerted effort to look at their wish list to see if we could oblige them. All of these points had to be combined with our vision of the Ty character - a whimsical and friendly Australian who always has enough time to help out his mates, drawing on not only what the world perceives as "Australian" but also on our distinct local culture and heroes. And we threw in more than a few Aussie jokes and slang along the way.
Making the Sequel is Twice as Much Fun
|An in-game screenshot from Ty 2.|
a sequel has quite a few advantages. Luckily, we knew that we were
going to produce a sequel before the first title was released. In
fact, we were well into Ty 2's production when the original
game hit the shelves. This time around, we already knew a lot about
the "Ty basics" - we already liked the character's look
and style, including his movement speed. We also knew how we wanted
to further maximize and enhance how his boomerangs worked as weapons.
Having these core components already figured out took away quite a bit of stress and allowed us to focus on other aspects of development. Additionally, the team that worked on Ty 2 was, for the most part, the same as those who worked on the first game. Since the majority of the coders already knew the Ty code, a lot of the "learning" period was taken out of the equation. The artists had a feel for the style of the game world, so there was less back-and-forth with the art director to get levels, props, etc. "right".
the team knew the hardware more intimately - the concurrent development
of Ty 1 on all three consoles resulted in robust cross-platform
knowledge that allowed us to push the technology and get a lot more
out of the systems. The only difference was the addition of the
GameBoy Advance version but we had set up a separate mini-team that
worked in parallel with the console team, and they were able to
take advantage of a lot of art assets created for the 3D game.
Throughout the development of Ty 2, we had a running joke that we made the first game on GameCube in about four megs of memory, as more and more memory was found "under the desk" of the GameCube engine coders. The advantages of using an in-house engine with "instant" support cannot be sufficiently expressed (though I know that the engine coders dreaded the sight of me towards the end of the project…). The engine guys did an awesome job as, despite the different amount of memory available, their work meant that we didn't need to make major compromises to get things to fit on the different platforms, which removed a significant headache that was present in Ty 1 (having to maintain so many different data files for each platform).
As a whole, I think the Ty team gained a lot from the development of Ty 2. As a result, we were able to create a sequel that delivered on many things we had set forth to accomplish:
- We know even more about how to structure a game that will appeal to younger audiences.
- We've used data streaming systems now, and we know what to avoid and how to plan the level maps to make the most of the technology.
- A task notification system has been created to limit the amount of time lost through miscommunication.
- We've adopted new production practices to catch as many implementation problems as possible BEFORE a task is begun (sounds silly, but it happens too much).
- All of our tools are more robust through another year of (mis)use.
- And last, but not least, we understand the dangers of excessive coffee consumption.
|Concept art of a lava spire from Ty 2.|
at the above, you might think that this means that making Ty
2 was an easier, smoother journey than the first game. Unfortunately,
due to a number of factors, this wasn't the case.
At the end of the production cycle for Ty 1 (late 2002), having just completed the PAL version of the game, it was obvious to us that there were several technological changes required for Krome to remain competitive and to be able to produce a game that contained the elements that the kids wanted. For that reason, work commenced on the Merkury2 engine (Merkury being Krome's in-house cross-platform engine). While many systems were carried across from the first version of the engine, there were significant optimizations made, including a change from being a frame-based system to a time delta system.
This change would allow any game made with the engine to continue running at a consistent rate across different frame rates (so, for example, Ty wouldn't have a slightly different jump path between the NTSC and PAL versions of the games). While this was an important and necessary change, it was also one that caused significant workflow interruptions - one side effect of this change was that a lot of game-side code also had to be modified, resulting in a few months of the game being broken. This meant that while the artists were working hard on new levels, they couldn't easily check their work. This removed two months of development time for the game, which was originally slated for an earlier release, leaving the team about seven months to finish the game.
During this period other work was completed that was of significant benefit to Ty 2. It's always been one of Krome's goals with Ty to create a game that truly represents the outback environment, with lush landscapes full of movement and noise. For this reason LETs (Living Environment Technology system) was developed. LETs is a system that procedurally places three dimensional trees, shrubs well… actually pretty much anything you want, onto a model using pre-set seeding values. These models are affected by wind, rustle when the player moves through them, etc. This was a significant improvement on the two-dimensional billboard grass that was present in Ty 1, and being able to place thousands of trees, shrubs, etc with minimal artist time that held up to close inspection was a HUGE timesaver, especially in the overworld that connected the different levels together.
In August of 2003, the development period for Ty 2 was extended to October 2004, but the scope of the game also increased. At this point we decided not to have any loading screens present throughout the game. So right near the end of 2003 (after nearly a full year of development on Ty 2), level streaming technology was introduced into the game. This change meant that more content could fit into a level by having discrete packages of memory load in and out as required. However, the down side to this change was that the large, sweeping levels we already built were not suitable for streaming, and so began a year long scramble to edit, modify and build new, more suitable levels. To further complicate matters, as we worked towards our last milestones, more and more streaming problems emerged as our internal QA department gave the streaming system its paces. If the player ping-ponged between triggers fast enough the system could be confused and packages of memory couldn't load and initialize fast enough, so delays had to be built in. There were also problems caused by some of the boomerangs in the game that could fly into areas that weren't yet loaded.
Many of the solutions we put in place to fix these problems were less than ideal. For example, the "airlocks" - rooms the player was sealed in until the next section was loaded - found throughout the game were just small rooms with hardly any props in them and were considered by many of us to be an eyesore. But with the amount of time left in the schedule they were an unfortunate necessity. The large overworld that was built, to connect all the different levels and missions, couldn't load and initialize fast enough without major popping, so chicanes and "slow-down" areas (sand and mud) were added. Though truthfully the sand and mud made the level look more interesting - all that asphalt got a bit monotonous. We added these areas about two weeks AFTER Beta, so things really did go right down to the wire. We made our gold master date by the skin of our teeth thanks to a huge effort by both our internal QA department and the dev team… and quite a few late nights and weekend work throughout the course of the project, (as well as some hefty content removal - at least 10-15 missions and two more kart tracks). This supreme effort resulted not only with us going gold on time, but we also passed hardware approval first time eight out of eight (Sony NTSC and PAL, Microsoft NTSC and PAL, GameCube with NTSC and PAL as well as GBA America and Japan). Once again, a HUGE thanks to all those involved.