Postmortem: Ubi Soft China's Music Up -- Summer Rainbow

Author: by Deng Yi Wen

UBI China had planned for a long time to make a PC game for the local market, but a number to factors kept the idea on hold. In January 2001, the right incentive to motivate UBI China to try a local project finally arrived. Shanghai Animation Film Producer, the leading local animation film studio, came to UBI Shanghai with the writes to “Music Up”.

“Music Up” is a cartoon television series about five teenagers who like music and organized a band, with the story build around what happens as the band grows up. Interestingly, the series itself was not as successful as expected but its associated merchandise -- books, cards, CDs -- are extremely hot among Chinese teenagers. A Music Up book series sold 350,000 copies in Shanghai. 350,000 in one city -- this figure became the most convincing statistic to push UBI Shanghai to make an investment in its first locally aimed project: Music Up -- -Summer Rainbow.

I joined the project in the middle of the development when the production was nearly frozen. Before me, there had been two project managers and one assistant who left the team after failing to work out the production problems. They started the project in March 2001, but work had all but stopped by November. Nearly 35 percent of the total budget had been spent, with little to show for the effort save a very short demo. Due to market considerations, perceived competition, and the need to time the release to coincide with the re-broadcast of the “Music Up” television episode, the game need to be on the shelf by August 2002 – and no addition funds would be forthcoming.

Characters from Music Up -- Summer Rainbow

I took on the project in December of 2001 and presented the beta demonstration at UBI’s offices at 18/F, Time Square, Shanghai at the end of April 2002. The master was ready in July, the gold master was sent off – just in time, on August 3 2002.

It’s not sure yet if Music Up is a hit with the local market or not, but the experience was extremely reward to the team and myself. Although we would have loved to lavished a long production cycle and bigger budge on our first game for the local market, Music Up – Summer Rainbow will make us more confident on the next local project.

What Went Wrong

1. Decentralization of the studios. In an effort to cut cost, the early managers of the project decided to farm out much of the production work to other studios. UBI China in Shanghai was responsible for the game design, quality control and final testing. The game engine was being built in Cheng Du, 2,600km from Shanghai and a center low-cost game production. The animation was taking place in Hang Zhou, 200km from Shanghai and known for budget animation studios. 3D backgrounds and audio dubbing had been sent to Beijing, home to China’s most diversified cluster of game studios.

The difficulty in working on such a spread out project is the communication lag. Although we built a server especially for the project, exchanging the documents by ftp, there was always a feeling of communication obstruction among the team members.

2. Poor game design structure. Music Up marked the first time that UBI China’s designers made a game independently, and the structure was changed continuously. The designers were constantly developing new ideas and the engine studio therefore always needed to change to accommodate the growing design. This, of course, led to increasing instability. More seriously, market demands meant that the game had to be complete for summer of 2002; time had run out for such changes but the designers still couldn’t help themselves from making more improvements.

3. No agreement on criteria for animation quality. The original design for Music Up called for top-shelf animation that would compare favorably to the Japanese games the dominate the Chinese market. However, the work we received from the Hang Zhou studio didn’t meet our expectations and was rejected several times. They declared that they had done everything according to the contract and the quality was in line with what we had paid. If we wanted higher-level quality, we’d need to spend more money. They disagreed with our rejection of the work and claimed UBI owed extra payments for rejected art.

The argument was very volatile and eventually led to a freeze of the production, with neither willing to concede a single point. When work stopped, Hang Zhou had finished 25 percent the contracted work (though our quality controller had accepted just 10 percent of the submitted files) to their satisfaction but received only 5 percent of the contracted payment.

4. Little money and less time. When I came onboard in December of 2001, I noticed that 35 percent of the total budget had been spent with only a 3 minutes demo to show for it. The product needed to ship in the summer of 2002 to catch the re-broadcast of the original series in the major cities and to avoid the strong competition from Sakura III.

It had also become clear that we would be seeing an additional investment in the project. A confidential sales report suggested that no single-player PC game would be sold well over the next couple of years in China, even Sakura II. Massively multiplayer games are expected to drive the market, surpassing even the popularity of pirated games.

With these two constraints in place, it was obvious that the theme of the production and development would have to be cost oriented rather that art first.

5. Expensive license. Obtaining the rights to Music Up consumed 25 percent of the total budget -- which conflicted with our philosophy of cutting costs as much as possible. If the money could have been spent on the engine, we would have realized more long-term benefits.

The marketing effect of building the game around the property was minimal, as the airing of the series didn’t favor of the games development and marketing at all. In return, it became a serious constrain of the production.

Most seriously, the music we bought with the copyright was not fully owned by the show’s film producer. The problem wasn’t found out until the production went to the final period. As a result, some music that the game designers most wanted was not available for use in the game. You can guess how disappointed the team was when we got the news.

What Went Right

1. Nice storyline and cartoon style design. As mentioned, “Music Up” is a cartoon TV series. The storyline needed to be updated to make a game, and game designers of UBI Shanghai, who usually make games for foreign markets, had the opportunity handle the design.

The storyline has changed. The five key teenagers are kept but another story was created and four more beautiful girls added. The setting was moved from Shanghai to Hai Nan Island in the most southern part of China, a place famous for romance.

Since all the designers involved are fans of Japanese cartoon stories and videogames, the game adopted a 2D cartoon style very much like Sakura.

The local marketing and sales team supported the design. They all agreed with the project and expected a success.

2. Difficulty package divided into manageable pieces. Music Up’s 2D animation was the game’s most critical component, and it was most difficult and disputed segment of the production. I separated the animation work into three parts, with the largest change being finding another team to take over the most disputed portion of the work -- the 2D backgrounds.

After the changes, the animation production went much more smoothly, though the argument over the quality never fully disappeared. We hired a professional animator (more on that later) to represent UBI China at the Hang Zhou studio producing the animation, and his presence made them more cooperative to make improvement on the disputed parts.

After successfully resolving the animation issues, other parts of the project went very well.

3. Part-time professionals invited to work for the project. Since our quality controller was not professional animator, we hired a professional animator to join team part-time and represent UBI onsite in Hang Zhou and placed him in charging of quality control. With the profession animator supervising the Hang Zhou team, they turned to be much more cooperative than ever and finished the target work on time every week.

On the advice of the game designer, we attempted to replace the problematic 2D backgrounds with pre-rendered CG backdrops. However, it was impossible to find a studio in the market to make nice CG backgrounds at a bargain price. In the end, we settled on a student group to product the work. Their asking price was acceptable modest, and we suffered only a slight loss of the quality. Considering that price was the most important factor in the case, working with a student team was by far the best alternative.

4. Tight schedule. In order to meet our goal of shipping the title in the summer, everything was arranged in a tight schedule. To keep on track, we arranged to closely examine our progress weekly and monthly, making very easy to find out were the production was lagging and push the team to finish it as soon as possible. In this way we were able to avoid any last minute surprises.

5. Cooperative team. A willing crew is the most important factor shipping a title on time, and we had a team who highly motivated by the chance to finally make a game for local players. I won’t say that there were any disagreement, but the team came together to meet our goal and get the game out on time.