and BTW, GPGPU is the reason Cell is not really relevant anymore, just saying.

Where did you get that info from? I've heard the tech in the Vita is really easy to develop for and it's clearly powerful while not too pricey for Sony or the consumer. 

The max resolution on Vita is 960x544. Uncharted runs at 720x408. Far from half the resolution and far from the 3DS's 800x240.

I don't think that development into Cell is worth the expense any more. GPGPU gives better performance and the research has already been done by AMD and NVidia. It's also easier to work with considering GPGPU cards have been around for quite a few years now.

I would really be interested in knowing how much two of these cell processors would cost for Sony because I wonder how powerful a PS4 would be with two of these and a decent APU from AMD. I don't know if the architecture would work together well, and the price might be too high for gamers, but it would make me happy to have backward compatibility, and seeing for my own eyes the difference between two cell processors vs CPU+GPU set up.

Not sure about the price, but 2 Cells together are practiacally QS21 BladeCenter - and they give 460 GFLOPS peak performance - that's lower than GPU inside A10-5700 alone (584GFLOPS). It just doesn't make sense not to use todays GPUs when they offer way superior performance.

The Cell would fare much better than the GFLOPS number suggests. The GPU achieves this with a bunch of brute force FP units - they have very limited memory access and can't work together on one thread or take normal-looking code. It's like you have a computer on every asteroid in the solar system and they can only communicate by flashing a single LED to each other.

Cell was much more CPU-like in memory access, many fewer threads were needed.

Cell was good at FP-heavy normal code. Unfortuantely game computing is INT-heavy and game graphics work best on a GPU. It's still THE best at what it does best, but it was never suitable for a game console.

Also remember the Cell was designed at 90nm. We will have 28nm in 2013, which will have ~24x the transistor density and so 24x the performance. Allowing for losses, Sony could put 16 Cells for the PS4 CPU and still be able to sell it.

Thought about this myself, many many times.

Whether the Cell has peaked or not, we just don't know. There are very few developers that can make proper use of it and those are Sony's best first party studios.

The easiest thing for Sony to do would be to continue to use the Cell for the PS4 and put in more ram and better gpu...etc. Would be great for PS3 backwards compatibility. Developers don't like the Cell though. It's too complicated. Sony got away with complicated architectures with the PS2, but the competition from Microsoft is too much this gen and they need to change their strategy.

So Sony is instead building a developer-friendly, affordable console that will last 7+ years again. This is the best thing they can do for themselves, and for us.

Which explains why PS3 exclusives look amazing despite its GPU being worse than the Xbox 360s.

The current source I am inclined to believe based on track record says PS4 is Spring 2014, AMD CPU+GPU.

The dual cell setup was never used not because of costs. The CPU itself had cost less to actually manufacture than the RSX inside it. The RSX was probably a reaction by other people in the company to the poor graphic API's that would come from using another Cell for graphics. Plus the RSX was designed in the PS3 a good two years before unveiling at E3 in 2005, so it was actually modified to run with Cell despite what people think. The biggest difference between Xenos and RSX is that Xenos was designed around early Direct X 10 API's which is why it's generally more flexible and easier to use, though it's not really more powerful than RSX.