QuakeCon 2005
John Carmack Keynote
The next major event at QuakeCon was a brief meeting of sorts to go over the upcoming movie based on iD's famous Doom franchise. Here, members from the studio walked through the premise of the movie with Todd Hollenshead from iD Software. Although this event failed to get the crowd overly excited, it was a relief to hear that the studio was remaining truthful to the franchise's past and that the movie would please even the most hardcore Doom fan. Fortunately, the event was short and John Carmack's following keynote was sure to not disappoint.
One of the most interesting aspects of QuakeCon is always the keynote speech by the legendary John Carmack. For the first time in three years, John managed to make it to the event to give the keynote in person. This year, the main topic of discussion was the next-generation consoles and how multi-core processors would effect the gaming industry over the next few years. Unlike previous years where John would whim philosophical about the most complex programming concepts you could imagine, this year's keynote was surprisingly easy to follow and digest.
Once the Q&A session began, we managed to ask John a question concerning the development cycle between the upcoming consoles and the PC. Without question, a hot topic of debate as of late has been the difficulty in programming and porting titles between the consoles. My question revolved around the optimization of titles for each platform and how a title simultaneously released on all platforms would compare with one another. As we know, there is an immense amount of resources necessary to get a title up and running on a single platform whether it be PC, Xbox 360, or PS3. Given the nature of the programming and development characteristics of each platform, it's not surprising to hear smaller developers being forced to choose either Xbox 360 or PS3 and not develop for both. Even the select few who do choose to develop for both run the risk of having sub-par titles as they lack the time and resources to optimize the titles for both platforms, and instead just get the games up and running.
Given the quality and polish we've come to expect from iD Software, I was eager to see what direction they would take. John's answer was surprisingly simple and direct, with no attempt to sidestep the question. With each console requiring a different approach to get the best performance from the hardware the approach taken in development would be critical. Here, John believes that the most effective programming model to produce the best results on all three platforms would result in developing around the PS3's CELL architecture. Although this would not be the most efficient method for programming for the PC and the Xbox 360, this is the direction which would likely result in the most polished title for all three platforms. However, exploiting the simplicity and ease of programming on the Xbox 360 is by far the easiest solution and is likely the route that iD Software will be taking. This model is much more aligned with the PC and will result in a much more tolerable development cycle overall. That is not to say that iD is neglecting the PS3 and is focusing on Xbox 360. Rather, it simply indicates that the PS3 development will require a bit more effort in the long run to ensure the same quality and gameplay we'll find on Xbox 360 and the PC.