Multiplatform vs Exclusive, what will sell more?

[kars]

Final-Fan said:

 

For example it is a well-known fact that the CPUs of the PS3 and 360 are radically different. I'm no programmer but it's obvious even to me that if someone uses every trick the Cell's 6 SPUs can be exploited to do it would be impossibly difficult to duplicate the results perfectly on the 360's Xenon. And, similarly, the Xenon's 3 fully-fledged cores are no doubt capable of feats that are a stone bitch to duplicate on the Cell with only one similar core, despite its 6 SPUs. And the rest of the hardware is different as well.

This is, of course, in addition to the simple fact that only one version of the game must be made, quality tested, etc.

While you are correct, that the CPUs themselves differ it is quite difficult to harness these differences. The best way to write software for the PS-3 is to start PPE oriented (there are severe problems if you try to debug parallel programs). When your program runs you can start to optimize by using the SPEs, but at the same time you can start to optimize on the Xbox 360 to use the other cores.

The problem is: such optimizations can become pretty expensive, compared to the better performance.

In fact the costs to port between Xbox 360 and PS-3 (and even the PC) are lower than they were in previous versions, while all the other costs (due to HD) went up significantly (better graphics need more details, which takes more time for the artists and so on).

These consoles have so much power that it is pretty easy to port between them, but due to the bigger complexity you will get more bugs and many of these bugs are plattform independent! So you won't really need 3 times more testers for three platforms, but you can attract up to three times as many customers.

I t is only a matter of cost efficiency and wether you like it or not, multy plattform games are on these plattforms much more efficient than exclussive games.