Postmortem: Frozenbyte's Trine
June 3, 2010 Page 5 of 6
3. Console game fail
For an independent developer, shipping a console game is not like shipping a PC game. This lesson has been repeated over and over by experienced developers, but being stubborn and having confidence in our abilities, we didn't pay much attention to it. The game was done and it was good, we felt comfortable with the console, early builds had met with great success at Sony. How hard could it be to get the final version released on PlayStation Network?
Trine was submitted to SCEE in the beginning of June and we considered it "done". We could not find any more critical bugs or anything that would prevent release. The plan was to release the game first on PSN and then a week or so later on PC, which needed some additional work on the menu system, mainly due to configurable controls.
When the PSN submission was returned with a bunch of bugs, we started to see that the plan would fall apart -- but little did we know that we'd have to pick up the individual pieces from the ground in a very tedious way.
The submission process was shrouded in mystery as all the communication with SCEE was handled by the publisher. Throughout the development, our relationship with the publisher had been difficult, and the submission process probably suffered from this as well.
Relevant information sometimes got lost along the way, despite (or because of) everyone on both sides working hard to get the game out.
Not only were we making our first console title, the publisher too was doing their first PSN title and the doubled inexperience really took its toll on the game. In addition to simple game-related bugs that needed fixing, a number of issues could have been solved by better understanding of the complex TRC document -- in other words, more experience.
Another group of bugs, such as wrong filenames or wrong program version numbers, could have been solved by a more experienced publisher, or by having direct control over the submission process.
We had to keep "supporting" the PS3 version until September. Fixing the bugs in each report was a matter of a couple of hours, but the mental stress started to build up.
The three to four months the game spent in QA caused a lot of anxiety and uncertainty on the whole team, most of whom had moved on to other projects by then. The team's, and especially the management's and the programmers', moods swung from the first submission's triumphant joy to annoyance to desperation to humorous disbelief during these weeks of constant waiting for new QA notes and final acceptance.
And it wasn't just internal -- many PS3 gamers were getting furious as the wait got longer with each passing week. Our long history of successful forum interaction with our fans backfired, when the estimated release dates we posted were missed time and again. When we realized the process could take longer, we became more vague with "no comment" posts, which did not do much to help.
We got more than our share of press coverage on websites all over the internet, and with each delay the user comments started to turn more and more frustrated, especially with the busy Christmas period and its slew of great AAA games coming up.
The PSN versions were finally launched on the 17th of September (EU) and the 22nd of October (US) of 2009 -- the one month delay between Europe and North America further highlights the submission and launch problems with the project, as the only major difference between the versions is the amount of languages -- Europe has six languages, North America only three. Both versions had been completed at the same time.
The situation remained much the same with patches -- the European version promptly got the patch that added proper Trophy images and made the last level easier, but the North American version had to wait over half a year until several non-technical aspects fell into place. It is much more difficult to get game updates released on the console platforms, and this was something we were not prepared for.
In hindsight we probably should have fixed many additional things during the submission process but after a couple of failed submissions we were too intimidated by the risk of breaking existing stuff and thus further delaying the game that we rather went the "let's get it out and patch it later" route. If we had known what we now know, we would have made more content changes to the submission builds and eliminated the need for patching as much as possible.
At the end of the day, Trine lost most of its PSN marketing and launch hype due to the delays, especially in North America. This, in addition to the vastly different release dates and inconsistent pricing across the platforms, undoubtedly affected sales greatly.
Page 5 of 6