What is the relationship between console horsepower and gameplay innovation? David explores the topic from a historical perspective.
In just three months Nintendo is going to reveal the first significant details concerning its next console to the world. The company has suggested that the new platform is going to feature innovations that will revolutionize the industry (likely significant additions or changes to what we currently refer to as the controller). Beyond this Nintendo has not revealed anything. Unfortunately, it is quite possible that, in the short term, they will reveal nothing more than the new hardware's technical specifications and name a few franchises that are intended to make up the launch lineup. In other words, the revolutionary essence of Nintendo's Revolution may remain a secret until much later. However, even without considering deliberate revolutions, gamers can still expect new things from the next generation of software, due mainly to the increased capability of the hardware. In this respect, the participants in the next round of console wars will be on relatively equal footing.
If we take a quick glance at hardware transitions in the past, the most notable changes have been improved graphics (excepting the transition from 2D to 3D, which resulted in dramatically different gameplay as well as graphics). However, with incrementally increased graphical capabilities and more powerful CPUs came new possibilities for gameplay as well. The first video games were played on immobile backgrounds (if there was a background at all). Not until the arcade game Defender did games begin to feature levels that scrolled in real-time to reveal what lies beyond the edge of the monitor. This feature quickly evolved into the expansive, colorful worlds of 2D side-scrollers like Super Mario Bros. As graphical capabilities increased again in the 16-bit generation, the size and number of onscreen characters increased, allowing for the design of games like Street Fighter II and Lemmings (the latter was eventually ported to the NES and Game Boy, but only by severely reducing the maximum number of onscreen lemmings from 100 down to a paltry 14). Even less innovative games such as Gradius III benefited from being able to display many more ships and bullets at once.
What we really want to look at is how gameplay has been evolving since the dawn of 3D gaming, and we only have one generation transition to observe. Interestingly enough, many of the changes that occurred this generation parallel those that occurred during the 16-bit generation. While the size of characters in 3D is relative (and thus meaningless from a hardware perspective), the number of onscreen characters has definitely increased as 3D hardware has gotten more powerful. Games like Rogue Squadron II, Pikmin, and Resident Evil 4 demonstrate this advancement nicely. Another major change since the first 3D generation is draw distance. Thick walls of claustrophobic fog have gone from commonplace to practically non-existent. Detailed physics simulation is just within reach on the current crop of consoles. However, rarely is it used to significantly change gameplay (as in Half-Life 2). Artificial intelligence (AI) has gotten incrementally better since the previous generation, but the changes have been subtle in most cases. Finally, as 3D hardware increases in power, split-screen cooperative modes are becoming feasible (at reasonable framerates).
This should give us a good idea of what to expect from the next generation of console games. Most of the things that were new or rare this generation due to technical restraints will multiply profusely in the next generation. There will be no shortage of polygons for drawing numerous characters. There will be plenty of CPU power for detailed physics simulation. Most developers will use physics for relatively superficial purposes, but some ambitious developers will certainly couple it intimately with gameplay, as Valve recently did in Half-Life 2.
Unlike physics and graphics, AI can be simultaneously a difficult feature to design and a computationally expensive one to implement, depending on the game type and the quality. The extra CPU power will allow developers to experiment with AI to a greater degree, but it remains to be seen whether or not additional CPU power will change the AI landscape. This situation is further complicated by the problem that extremely realistic AI may not improve (and could even hamper) gameplay. For example, extreme realism can make things too difficult. Good AI should also be somewhat unpredictable, but if gameplay designers cannot predict enemy behavior, how can they design a fun game? One new option in regard to AI will be using some of today's more complex AI in conjunction with a much larger number of individual game elements.
Cooperative modes will not necessarily become extremely common in next-generation software, despite the increased processing power, but they should be at least more common than they are now. The problem is that many of the current barriers to including a co-op mode will still exist. In a split-screen co-op game, the number of triangles manipulated per second can easily double, if detail is not removed from a scene. This can be a major problem for any hardware that is being utilized fully. Fortunately, pixel-shaders, which will take up much of the graphics processing time in future software, can be toned down for co-op games if necessary. However, other hidden costs must be considered as well. AI must react differently if enemies must deal with two or more players. This forces developers who want to include a co-op mode to plan for it early. The flipside of this situation is that online cooperative modes could become more feasible. Technically, online co-op requires the addition of optimized network code, multiplayer AI, and a touch of extra graphics power to draw two or more players instead of just one. The biggest factor is the network code, which may be encumbered by next-generation physics and AI simulation. This still requires planning from the beginning to be fully effective, but with online gaming expanding every year, hopefully more developers will consider adding some sort of co-op mode to their design documents early on.
Another thing we can expect from the next-gen is more gameplay involving light and shadow. Currently, very few games (Splinter Cell is one) rely heavily on light and shadow to create gameplay scenarios. Increased GPU power will make this an option for less technically proficient developers, and we should see real-time light and shadow used in larger environments.
And lastly, the number and scope of massively multiplayer online games will likely increase significantly in the next generation, due to the increasing number of online players and the increasing prevalence and bandwidth of Internet connections. Whether or not Nintendo plans to get involved in that revolution remains to be seen.
Clearly, Nintendo's stance that console gaming requires some sort of revolution to maintain the interest of gamers is somewhat premature. While there is a lot of repetitive game design in the industry, there is still plenty of innovation in certain corners. Even 2D gaming, which has been around for more than thirty years, continues to attract the attention of both developers and gamers. Yet Nintendo is ready to bury the current paradigm of 3D gaming and move on to supposedly greener pastures. We can only hope that the design of their next generation console allows the current style of 3D games to live alongside the new. There is still much fun due gamers from mere evolutions of current genres and many classic franchises that have not yet made the leap to 3D. Whatever happens, Nintendo's course is now set.
Please visit our Talkback forum to discuss this editorial with other readers and the author.