I just posted this in the Sony Upheaval Thread but it probably belongs here - Please don't inflict reading it twice on yourself
45% Opinion-65% Sure
The PS3 may become nominally profitable but I doubt that it will ever begin to make up the lost ground and it will not last ten years or lead directly to its successor because I think the cell architecture, while a brilliant solution to get that kind of performance today is technologically a dead end. Ultimately the games created for it will likely not be BC on the PS4 nor will the tools and programming skills for it transfer forward. This will have to affect its attractiveness to 3rd party developers.
Fact
The danger of trying to leap too far ahead of mainstream technology too soon is that when the mainstream catches up - you're dead. I'll use an example that I am intimately familiar with. Up until fairly recently, the only way to get a high resolution digital microscope camera to operate at full resolution (1024x1024 and up) at full video rate (30fps) so that you didn't have visible lag was to use special one-off proprietary cards and wide parallel bandwidth cables. These systems worked well eliminating the lag (slow refresh) that otherwise made viewing unpleasant and focusing difficult. The downside was that they were very expensive, not easily portable because of the board, and because these specialized PCI cards were by small manufacturers, compatibility issues were not infrequent, special drivers were required that restricted your choice of OS, and only software especially designed to recognize them could be used, except with a slow and clumsy TWAIN interface.
Be that as it may, they were the cat's pajamas, the instrument you had to have regardless of price right up to the exact moment when the first CMOS/IEEE1394 (standard Firewire) and soon thereafter USB2 cameras came out. They did the exact same job but cost about 20%-40% as much, used a standard interface, often required no driver and could be moved freely from computer to computer. At that moment the sales of the proprietary cameras went to 0%, support went to 0, and you could hear the sound of $10,000 cameras hitting trash cans. Instant orphaned technology.
40% Speculation- 60% Sure
I don't know the exact time but it'll be a lot less than 10 years. I don't know the exact form, Fusion, NVIDIA, INTEL, or Open GL but essentially everyone is working on the GPU based parallel processing computer. They exist now but not yet in a form or at a price for video gaming. Some people are not sure if they can ever completely replace the general purpose CPU for non-parallel computing functions, math, word processing etc but there is no question that it is the future of graphic intensive massively parallel processing. Game consoles would be an almost ideal application.
In much less than the 10 years that some think the PS3 will be leading edge, it will in fact be vastly outperformed by an inexpensive standard GPU processor than can be programmed like a standard PC, meaning essentially, no more ports, no more multiple versions, and competition between platform will be strictly on the basis of exclusive IPs and GAMER INTERFACE. Since video standards will only go up at very infrequent intervals, these consoles could have extremely long lives with software and firmware upgrades.
100% pure Speculation
Obviously hacking and pirating could be a problem but not necessarily an insolvable one. Games could, with memory being so inexpensive be in the form a high speed cartridge, perhaps a solid cast one to defeat disassembly or even now that most consoles are internet ready, online activation. Otherwise software and hardware could become separate entities like games for windows, where it doesn't matter what you play them on.
This solution would probably fit MS fine but I'm not sure that Nintendo would be willing. I'd love a dual system though where Nintendo first person came on a cartridge but you could also play all the other titles. That would seem to screw Microsoft but if they stuck to traditional controls and Nintendo went totally motion, the separation would probably be about what it is today. What about Sony. If they stuck with the PS3 for ten years, they probably be a tiny niche. If they jumped into the fray, they could possibly be the casual-hardcore bridge machine with a dual control system.
When will a great commodity priced GPU HD processor be ready to go? Beats me, but if I knew when Nintendo was planning the Wii2, and probably if I knew the date that MS planned the 720, I think I'd have the answer because I think this is why they took the routes that they did.
