4. Kismet and Animation

We had problems with Kismet, which was used by designers for the prototypes. Their job was well done, but a problem appeared in the final version of the game. We saw real monsters built from the blocks, and gargantuan links between independent elements of the game. It was a source of extreme errors.

We also had to deal with the framerate and drops in performance. For example, the reaction to the attack was associated with face and body deformation, facial expressions, blood or bruises. Problems ended after we transferred those events to the Unreal Script -- programmers rewrote what level designers had created in Kismet on the scripting language. It gave us better control and improved performance.

We also encountered serious problems with the animation. Perhaps we didn't spend enough time on research and training, because part of it was incorrect. Particularly troublesome was the distance between players. We recorded three different distances for each hit, so that devs could afford to choose the best one.

However, we still had the unusual situation when the boxers began to attack at the same time. We had to use a couple of tricks to overcome this including collision detection, inverse kinematics and selection of the most appropriate action.

5. Crunch Time

The decision to cooperate with external companies was correct, but we found out that reporting errors and bug fixing became much harder. Direct contact always shortens the path and speeds up the process. It was the right decision, but passing the data to the new programmers who later joined the production almost killed our server on the same day when we had to finish a valid build!

Like most developers, our team was working flat out at the end of the project to deliver on time, as Apple had indicated they would possibly get behind the game in a big way. During this time, one of the rooms literally turned into a repository of energy drinks and local restaurants provided meals on a daily basis for those working extra hours. The ambitious nature of the project put massive pressure on the team, and each developer was aware of the need for it.

There is no simple recipe for a solution to this problem. We can't pretend we found one either -- yet -- but we will change few things in the future. To name but the smallest one, for example, as we experienced occasional delays due to lack of complementary equipment -- graphic designers required additional monitors and had to wait for delivery, so we'll surely optimize the work stations earlier next time.

In the case of a tight schedule, every short downtime can cause people to stay longer. For our next high-budget title we will spend more time planning, researching and will invest even more resources into the project preparation. For Real Boxing, we knew that we were working on a triple-A title, but that's easier said than done.

It is certain, however, that every team needs to rest and regenerate after such a hard work, so everyone got additional paid vacation after the release of the game and began to receive overtime. We also prepared a huge “Thanks for the hard work” banner for the office.

Final Verdict

We feel that with Real Boxing, we achieved a victory, one that was confirmed by reviews and gamers alike. The investment in quality shone though, and we were awarded the coveted Editor's Choice by Apple, and received almost universally great reviews from all the major media that we targeted.

We also learnt a lot of lessons, as developing high-end games for mobile devices is new territory. Real Boxing has by far and away been the biggest venture for our studio and brought invaluable experience. But this is not the end. We plan on supporting Real Boxing by providing updates, releasing new content (arenas, boxers, items), and expanding the multiplayer mode. We're also looking at the possibility of releasing our game on other platforms. Of course, Real Boxing is just the first of a few high end titles we've got in the works, so watch this space!

Data Box

Developer: Vivid Games
Publisher: Vivid Games
Release Date: 15 November 2012
Platforms: iOS
Number of Developers: 25 full-time, 8 subcontractors
Length of Development: 5 months
Budget: $300,000
Development Tools: Unreal Engine, Scaleform
And: More than 500 energy drinks, few boxing training sessions, 25 pairs of boxing gloves, about 200 dinners, one punching bag, a few thousand coffees, one big Thank You banner for our team.