Yeah as much as I’d love to get into an in-depth discussion about this, you’re not going to get much detail in comments about it because of NDAs etc.
I think that the conception of the PS3 being harder to program for is a combination of the tools, the architecture, and the maturity of the console. One of the biggest strengths of the 360, development-wise, is that it’s a Microsoft product – with that comes all their expertise in development tools and documentation. The 360 has been around for that much longer, and people have learned what works and what doesn’t – the PS3 is still playing catchup in that regard. Add on to that the radically different architecture of the PS3 compared to any other multithreaded hardware system and you get a very steep learning curve. It’s natural that it will take a while for titles to take full advantage of the PS3 – when PS3 devkits have been around as long as 360 ones are now, I think it will be a different story.
Comparing statistics is largely irrelevant; each has its own strengths and weaknesses, and what works well for one won’t for the other. I suspect that when it comes to PS3 development, a large part of the problem is that many teams are leading with 360 and leaving PS3 as almost an afterthought, to be optimized at the end. That simply won’t work, and could well be why we’re seeing PS3 versions of games delayed while the 360 version is released. The game engine’s design has to be considered from the start with respect to the PS3’s architecture, which requires a very different approach in order to make good use of its strengths.
At this stage I should probably add that this is all my own opinion, and doesn’t reflect that of Radical :)