I'd have guessed 2010, given Microsoft's love for short upgrade cycles, but they'd have announced it by now if they were going to release it that soon, so that's out. Maybe 2011.

They'll try to launch first again to be first again and have the most reliable console again!!!
...Sorry, what does it mean "It didn't work"? And most important, WHO'LL TELL BALLMER???  

(A) The Cell is actually not a flawed concept at all, nor is it the reason the PS3 costs so much (Blu-Ray is).  Cell chips aren't any more expensive than Xenons to produce, and frankly, IBM advances its manufacturing process faster for the Cell than they appear to do for the Xenon -- the 45nm Cell PS3s are right around the corner, as I understand it.  45nm Xenon/Xenos combo... next year sometime.  I don't even think they've managed to make a Xenon/Xenos 45nm prototype yet, whereas the Cell has had a 45nm prototype for over a year already, and the RSX is probably not far behind.

(B) You're wrong.  That, and a big chunk of the SPEs major awesomeness comes from their ability to process quaternion and vector math really quickly, and en masse -- namely doing animation processing, which is Huge in a large number of game genres, with a capital 'H'.  See the following genres: FPS, Sports, RTS, Action/Melee, anything with large numbers of detailed 3D characters onscreen at once.

(C) Such a machine would use way more transitors than would be affordable, and would actually cause (A) to be true.  The Cell is great because it is a huge amount of processing power for relatively few transitors... i.e. its cheap for what its capable of, from a hardware standpoint.  Its expensive to develop for, because the hardware is very nonstandard.

You could claim that Sony made a mistake, by including Blu-Ray in  the PS3, and thus, hiking the price.  The initial runs of the Cell were very low yield, and so they did cost alot in the beginning of the PS3's lifetime, but it was only about 6 months before Cell production was very efficient, and the Cell's cost dropped to that of its contemporaries.

Sony paid a lot to win the format war... but they stand to gain much more.  The PS3 is far from being a failed platform, and Blu-Ray has succeeded as well.  Its inevitably the next major disc format, and that will pay for years and years to come.  If the PS3 ever exceeds the sales of the 360, and in one year's less time*, and Blu-Ray wins the format war*, Sony will have trumped this generation in a huge way.  The only things the PS3 has failed to do is to catch, or surpass, the 360, in less than 70% of the time the 360 has existed, which is a nonsensical idea to say the least, and capture the Blue Ocean, which Sony obviously wasn't targetting (although I imagine they, and MS, want to), and Nintendo now owns.

* Already happened.

i reckon 2012, just think how ridiculously expensive it will be though , newer larger games will either need blu-ray, or some brand new disc format, or one big mother fucking hard drive , it will need to make hardware that's better than the playstations 3's cause there would be no point in making a console equal to ps3 , so i see a huge ass price tag coming , get saving now and you might be able to afford it on release

The Cell isn't flawed as a concept -no more than any other kind of processor, anyway- but Sony failed to release proper tools for programming the thing. The tasks it's well-suited for don't have to be hard, but they haven't put in the effort to make it easy for the developers, and that has killed every other console that tried to deviate from the typical architecture in the past.

It would have killed the PS3 by now, too, except that the HD market is too small. It can't sustain third parties with only one console, so devs have little choice but to go multiplatform: the cost of porting is still less than the money from another console's fanbase, and while it wouldn't be enough to be worth bothering with in a healthy market, right now it's essential for keeping third parties afloat. Thus, the PS3 still gets games.

@Procrastinato:

I would say the major flaw in the Cell relates to the internal memory cache architecture of the SPE's. If the Cell used a similar shared model to Larrabee it would have been far easier to design programs and achieve closer to the maximum theoretical performance. Unfortunately changing this architecture would break the backwards compatibility with the PS3 most likely.

hopefully never

September 2011, 6 years after the 360 launched, and it will be an awesome console

6 year is a bare minimum life as high end console for this gen, unless the producer followed Nintendo's way: initial investments wouldn't be recovered otherwise, or, in the best case, net total life profits would be so small to give a very little justification to billion $ investments. If they plan to kill XB360 short after XB720 release, instead of following Sony's way to keep the old console as entry-level offer, they'll likely need an at least 7 year lifespan for XB720.
Obviously, if MS' main goal is to harm Sony and not profiting too, this doesn't apply, but I have some doubts, Sony itself prevented PS3 from being the danger MS feared, as too little RAM makes it not viable as home/living room computer, right now MS should fear more Nintendo, and the tactics it used against Sony are already proven ineffective against the current market leader.

If Microsoft lauch next XBOX BEFORE 2012 will literally kill PS3, I am sure of that... and no PS3 is comming until 2016..

So... I think there´s no XBXOX 360 or PS4 for sure, because next step isn´t a new console but make console internal to others hardwares such as TV´s and HTs...

let´s goooo