Most of this post supports my point.  You're right that things like automatic load balancing don't "just happen" because the SPEs can only work on their local store -- all the more reason to think carefully about how your application uses memory in the design phase, and how it will give work to the SPEs.

I'm not saying it's going to be easy!  My point is that you can make it less difficult by thinking about the platform before you start writing code.  I'm not saying you should just code the whole thing, using all the SPEs, press run and hope for the best; that's foolish.  I would take the modular approach, write and test portions of the code at a time until it all comes together.  Obviously there will be room for tweaking further down the development pipeline, but it helps to have an idea of what you can achieve before your engine is close to being finished.

For the most part third party developers do not care about quality ...

The reason you will get better quality Wii titles from third parties than what we have received so far is that as the quantity of titles increases and you receive greater support from high quality third parties you end up having a lot more high quality third party games.

Trust me, after E3 people will think of it as the "Nintendo Show" because the amazing performance of the DS will lead it to have the most high quality games announced for it, and the increase of support we have heard about (and been seeing)  for the Wii will lead it to have far more announcements than either the PS3 or XBox 360. Even though Nintendo doesn't participate at TGS I expect there will still be a strong showing there.

Regardless on whether there is a shift away from the PS3 the Wii is getting more support and this means that there will be a lot of announcements because (unlike a lot of PS3 projects) we have no knowledge of these projects.

 You should say it differently: how it might work in connection with SPEs.

If I look at my experience, it is quite hard to determine the really processing intensive part of your program beforehand. If you put less used code to the SPEs you won't achieve drastic changes with them, that's the problem.

Yes, you might try to keep the SPEs in mind but it is simply useless to put much work into this part before you have a hlaf finished product. The real processing intensive code will grow in parts, where you didn't expect it to be, because for some difficult parts you found rather easy solutions, while some minor problems suddenly became big ones.

Most of your planning won't survive the implementation phase. In fact some of the code that you designed for the SPEs will suddenly be slower than the integrated solution.

The SPE concept has the problem, that it depends on structural parameters, not functional ones.  But the structures evolve and change much faster during implementation, than the functions. In fact, if you want to protect structures you will probably get slower code, because the code around these structures will have the tendency to become more and more complex.

Yes, it is possible to design certain SPE functions beforehand, but I would presume that this code won't be used as often and as efficiently as you orginally planned. To really change the performance of the processor as a whole you have to find and export the critical functions. Cell programmers have to know Amdahl's Law by heart.

@HappySqurriel

I think I read at Neogaf that Nintendo will be at TGS this year because Dragon Quest to DS.

@Shane

Again price or development time has nothing to do with a game being AAA or A. It is all about the quality of the game. The price might mean that the company doesn't dies if the game flopps, but it has nothing at all with the quality of the game.

Do you or do you not see LittleBigPlanet as a potentional AAA title? This game is developed with a small bugett so far. It doesn't use all tha power of the PS3 but is for me one of the games that is on a potentional must have played list.