Tetsuya,
There may be a level where you can say a game is a "hit" or not... I'd say if a game sells 1M in "the Americas" or 300-500K in Japan, that game is a very significant hit, and if it manages 2M in America or 1M in Japan, its an absolute blockbuster; a "killer app," if you will.
But that says NOTHING about whether the game is "successful." Prof. Layton on DS may manage 700-800K sales or more in Japan. For a game of its type, its a highly impressive performance.
But if Dragon Quest 9 came out and did 1M in the first week in Japan, it would be a poor performance. The mark for DQ games is not simply being a "hit," but being the biggest hit of the year and one of the biggest of the generation, with 1.5-2M+ first week, and 3.5-4M+ lifetime.
Different games have different marks for "success" and "failure." It has to do with expectations, which are based on the country, the time of year, the install base, and precedents set by past games.
"[Our former customers] are unable to find software which they WANT to play."
"The way to solve this problem lies in how to communicate what kind of games [they CAN play]."
Satoru Iwata, Nintendo President. Only slightly paraphrased.
