How does a console finish its time on the market once its successor launches? Does it linger on for years, as sales slowly wind down? Does it bow out quickly, leaving the spotlight for the new generation? The answer to that question depends on a lot of factors. As I'll show you below, some platforms like the Nintendo 64 and GameCube departed within a year of being made obsolete, while others hung around far longer. And starting this winter, both the Xbox 360 and PlayStation 3 will no longer be the flagship consoles for their respective manufacturers. However the market today is not the same one into which they entered, and there is a chance that their passing is an opportunity for new systems to break into the console market. Sony and Microsoft and Nintendo are still working the console market from their traditional assumptions -- and those assumptions may not be valid much longer.

The Nintendo transitions

Before I get to that, however, let me show you what has happened to Nintendo's console shipments once a new system is launched. Looking at Nintendo's earnings releases, you can piece together the following data for its past three console transitions. Once the GameCube launched in late 2001, only about half a million Nintendo 64 systems were shipped, and no more systems delivered to retailers after March 2002. The same kind of transition happened between the GameCube and the Wii, but over a longer period. Between the end of September 2006 and the end of March 2008, Nintendo only shipped about 540,000 GameCube systems. In both cases, post-successor sales of the older console was a tiny fraction of overall sales. However, the situation with the Wii after the Wii U launch has been quite different. From October 2012 through June 2013, Nintendo has shipped an additional 2.9 million Wii systems. This includes the Wii Mini, which first began shipping mere weeks after the Wii U launched, an indication that Nintendo still sees life in the aging little console. One explanation for the longer Wii afterlife is that Nintendo hopes to send the cheaper console to countries outside its traditional territories which haven't yet been tapped. Or, Nintendo could simply be responding to the remaining demand for its older console – the Wii is still outselling the Wii U in the U.S. for example – and helping to pad its bottom line each quarter. Likely the real answer is somewhere between these two. One way to visualize how Nintendo has phased consoles out as new ones became available is to track the company's annualized (trailing twelve month, or TTM) sales rates for each system. That graph, for the last four Nintendo consoles, is shown below. (Another view of the same data can be seen here.)

The PlayStation transitions

Turning to the other Japanese console maker, Sony, we can see an entirely different picture. Here are the figures I have for the transition between the original PlayStation and the PlayStation 2 and then between the PlayStation 2 and PlayStation 3. I hadn't thought about these figures for Sony before working on this column, and I was frankly quite shocked. In both of the previous transitions, Sony shipped approximately 30 percent of its older console's total units after its successor launched. In the case of the PS1, that turned out to be 30 million systems over the course of over five years. For the PS2, that came to 50 million systems over the course of more than six years. If the PS3 were to perform similarly, its final sales total would come out to 104 million systems, just a few million past the final total for the PS1. Let me give you the TTM graph for Sony's hardware similar to the one I made for Nintendo. (And another view is here.) But will the PlayStation 3 really sell for years after the PlayStation 4 launches in November of this year? I don't think that's altogether obvious, because a console similar to the Ouya or Valve's Steam Box could disrupt Sony's console afterlife. These kinds of mobile-based consoles will have advantages that Sony simply does not with its console. The PlayStation 3 is designed around the custom Cell processor, a custom Nvidia graphics processor, and a hard drive. I believe there is a floor of around $150 per system below which Sony will never be able to get their PS3 production costs. It has taken seven years to get to the $200 level in the U.S., and that system has less on-board storage than the original $500 20GB model had in 2006. While Sony continues to battle the consequences of the hardware decisions it made back in 2005, the newer consoles will be built around the reality we have today. Using low-cost CPUs and GPUs, a modest amount of on-board storage, expansion in the form of SD cards, and cloud services to provide software on demand, an Android-based console starts at a far lower base cost. As mobile-based hardware and operating systems migrate to these novel consoles, a whole army of developers and a slew of well-known, newer game franchises will follow. Even big publishers like Electronic Arts are now extremely active on mobile platforms, and it wouldn't surprise me to see all their big franchises appearing on a console with the right pedigree and userbase. Moreover, the hardware and game engines have been ramping up their sophistication extremely quickly over the past few years, following the kind of growth curve described by Clayton Christensen in his work on market disruptions. That means that a new console could come in and compete with on the features like graphics and sound that have traditionally been PlayStation's strengths, and still undercut Sony on price.

Disrupting the traditional console transition

That competition could, of course, have consequences beyond trimming Sony's hardware sales. All the additional PS3 software revenue that Sony could expect over the next several years, both in terms of sales and licensing fees, will likewise take a hit. Of course, this is all speculation. The current competition, like the Ouya, simply isn't a real threat to Sony. However, it would only take one surprise announcement from a big company like Apple or Amazon to change the landscape completely, and it seems certain that Valve will eventually bring its Steam Box to market. I'm reminded of the saying that generals always prepare to fight the last war. Microsoft was soundly beaten by the Wii when it was unprepared for the novelty of Nintendo's motion controls – and now the Xbox One with Kinect has the most tightly-integrated motion controls any console has ever had. Sony was criticized for its more challenging hardware and high costs, and is now praised for the developer-friendly PS4 and lower entry-level price. But their business models are still mostly built on the presumption that there will always be a traditional console market. It will just take a single successful outsider console to change everyone's perspective. [Note: There is some fine print attached to the Sony data I've presented above. Prior to April 2006, all the data Sony provided was production shipments – that is, systems produced by Sony and shipped to warehouses but not necessarily sold to retailers. Starting in April 2006, Sony began reporting sales to retailers. I believe over longer periods of time (on the scale of years) these two figures roughly track with each other, and so I've not bothered to separate them. Finally, for the last five quarters Sony has regrettably combined PS2 and PS3 sales figures in its quarterly statements. I have used my own personal estimates for each console's share of each quarterly sales figure in those cases.]