I was using the DS as an example because it goes against the idea that people bought a system just for a gimmick.  It's hard to argue that a touch screen was very exciting tech, even in 2004. Despite this, Nintendo's software on the DS performed very well.   People didn't buy the console because it was like "OMG IT HAS A TOUCH SCREEN" they bought it because they saw software that looked fun. Same thing with the Wii.  

Wii sales increased because the Wii had games that were more appealing to a broader audience.  Those people bought the Wii because they wanted to play the games they had for it, for instance games like Wii Sports or Wii Fit.  Once they had the system, they bought games like Mario Kart, Mario Party 8, New Super Mario Bros Wii, Animal Crossing, Kirby's Epic Yarn, Wario Ware Smooth Moves, Donkey Kong Country Returns, Brawl, and so on so forth, because they thought they looked like fun.  

You could argue that those people just bought the games just to shove something in the Wii, but I don't think the kind of audience the Wii has is the type of audience that has a compulsive need to buy games just to play them.  Besides saying that "Nintendo fans buy Nintendo games because they have no choice" doesn't nearly account for the staggering sales of games like Mario Kart, NSMB, Pokemon, Mario Galaxy, Nintendogs, Wii Sports Resort, and so on.  When we're talking about sales of 10-20 million, "there's nothing else on the system" hardly seems like an adequete explanation.  

And of course, there are TONS of loyal Nintendo fans who followed Nintendo throughout the years.  For a game like Brawl or Skyward Sword, you'd have to think that the lion's share of sales are from devoted fans.

Fact of the matter is that Nintendo games sell very well because Nintendo is very good at producing games that people feel are worth $50 (Nintendo games generally take a long time to drop from that price).  Sony's games don't sell as well because Sony does not make games that a large amount of people think are worth 60 dollars.  Sure, third party competition would make some difference, but it's not nearly enough to explain the large discrepancy.

I'm confident this game will do 5m over a few years.

To all those mentioning sales discounting it doesn't worry Sony and other publishers if retail discounts the games down the track since they sell the games at a upfront wholesale price and doesn't really matter if all or half the stock is sold at RRP since they have had their cut .

PS3 games always have great legs, espicially compared to the xbox 360 version of the same game or even exclusives.

Fix the first week sales for all those games and you might have a good debate. But 25k for Heavy Rain when it actually sold over 400k is a pretty damn big discrepancy.

UPDATE

Heavy Rain FW sales changed to 413k
Uncharted FW sales changed to 108k
Resistance FW sales changed to 194k
Motorstorm FW sales changed to 221k

When he adds the tie ratio after the first week sales we can talk about something. The first week figures of those 2006 and early 2007 games are deceiving.

Agreed, the lifetime multiplicator would be lower. But the first year multiplicator would be the same. In fact, it would probably be higher due to a higher install base.

I'm not sure what to do with Demon's Souls. It launched in Feb in Japan selling 38k units FW. Then it launched in USA in October selling 53k FW. And it launched in June the next year in Europe selling 30k units. All together, that's around 120k.

But by the time it launched in USA and especially in Europe, it certainly wasn't a new Ip anymore. The game had already gained several months of hype much before it's release. It didn't face the circumstances of a true new IP. Therefore, I'm not going to give it the full 120k, since it wouldn't have sold that much if it launched worldwide at the same time. Instead, I'll give it the 90k

I'm making calculations. Calm Down!