I read this a while back, and was scared a bit at first, but then someone in here @ AI straightened me out a bit.
If anything it'll probably benefit Apple for IBM to be designing, and manufacturing the CELL processor for Sony. I'm sure they will learn quite a bit from the technology, and in turn use translated PPC similarities to accomplish more with the IBM architecture. I think IBM will become a more proficient, and intelligent processor manufacturer because of it.
I read this a while back, and was scared a bit at first, but then someone in here @ AI straightened me out a bit.
If anything it'll probably benefit Apple for IBM to be designing, and manufacturing the CELL processor for Sony. I'm sure they will learn quite a bit from the technology, and in turn use translated PPC similarities to accomplish more with the IBM architecture. I think IBM will become a more proficient, and intelligent processor manufacturer because of it.
I read this last week, and laughed at his sugestions. First off CELL is designed to be used in consoles and possibly other consumer electronics. The target price for these systems will need to be $300 at most, knowing fully well that in 12-24 months it will be selling at half that price. These chips will need to be CHEAP and due to that will probably not be as high a performance chip as the current level used in even the lowest cost PC's. When the Cube came out (~400 mhz G3 based chip) we were moving to, or getting ready to move to 466-733 mhz G4's for the PM's and the iMac was at no less than 400 mhz-600 mhz G3's ( 2001, not sure when the Cube came out but it was no earlier than that), and that was after a 18 month lag in G4 speed improvements.
If Cell makes it on scheduele, and has as many cores as rumored, I would expect each one to be fairly low speed and transister count to maxamize the number of usable chips in a production run. Don't forget that for the chips to be usable it will each core will have to pass the tests at the target clock speed, and there is no guarentee that each core will do this. Therefore they will be clocked welll below the "maximum" potential clock speed and have as simple a design as possible to maxamize usable chips. If not Sony wont't be able to ship enough PS3's and may loose market share to Microsoft. I don't think they will risk that on to complex of a system that may limit their ability to actually deliver the systems.
Manufacturers loose BIG money on console sales. They gain it back with games. Processor cost isn't as much of an issue as it would seem.
Being that video game sales are second only (and barely) to the movie industry (and growing) for entertainment sales I don't think they are concerned about the price so much for this new generation console either. Billions, and billions.
The other thing to note is that IBM is also designing, and manufacturing the new processor for the XBOX. Sony got into bed with IBM with CELL for the Playstation first. Not that it necessarily means anything in big business anyway, it was just something I thought to add to pot, and stir.
Actually Sony has wide ranging plans for Cell that extend well above consumer entertainment devices. Here is an example of higher end non-PS3 hardware Sony is working on:
There is actually quite a lot of information on Cell on the Net from patents, public talks by Sony people, and probably some people breaking their NDAs. From everything I've read so far, a Cell powered PS3 is going to be a monster performance-wise. Although it is difficult to tell which information is PS3 specific and which is referring to other future Sony hardware. It is fascinating stuff for anyone with even minimal chip tech knowledge. I can't wait to get my hands on a PS3 dev unit.
I would assume Apple has had talks with IBM for some time now about the possibility of something similar to Cell type chips in Apple hardware in the future.
Nintendo has always been near the break even point on their console hardware. Sony hit the break even point about a year after the PS2 was released due to their ability continually decrease their manufacturing costs.
Even if Cell isn't officially Power/PowerPC-based it's highly likey that it is. Sony is a recent PowerPC licencee and IBM is talking about their 440 processor more and more as a cell design, especially for their BlueGene/L-supercomputer design.
"A lot of people have excused this on the grounds that the move to 90-nanometer manufacturing has proven more difficult than anticipated, but I don't believe that."
What's not to believe? This guy is crapping out of his mouth for this story.
Comments
If anything it'll probably benefit Apple for IBM to be designing, and manufacturing the CELL processor for Sony. I'm sure they will learn quite a bit from the technology, and in turn use translated PPC similarities to accomplish more with the IBM architecture. I think IBM will become a more proficient, and intelligent processor manufacturer because of it.
Originally posted by onlooker
I read this a while back, and was scared a bit at first, but then someone in here @ AI straightened me out a bit.
If anything it'll probably benefit Apple for IBM to be designing, and manufacturing the CELL processor for Sony. I'm sure they will learn quite a bit from the technology, and in turn use translated PPC similarities to accomplish more with the IBM architecture. I think IBM will become a more proficient, and intelligent processor manufacturer because of it.
I read this last week, and laughed at his sugestions. First off CELL is designed to be used in consoles and possibly other consumer electronics. The target price for these systems will need to be $300 at most, knowing fully well that in 12-24 months it will be selling at half that price. These chips will need to be CHEAP and due to that will probably not be as high a performance chip as the current level used in even the lowest cost PC's. When the Cube came out (~400 mhz G3 based chip) we were moving to, or getting ready to move to 466-733 mhz G4's for the PM's and the iMac was at no less than 400 mhz-600 mhz G3's ( 2001, not sure when the Cube came out but it was no earlier than that), and that was after a 18 month lag in G4 speed improvements.
If Cell makes it on scheduele, and has as many cores as rumored, I would expect each one to be fairly low speed and transister count to maxamize the number of usable chips in a production run. Don't forget that for the chips to be usable it will each core will have to pass the tests at the target clock speed, and there is no guarentee that each core will do this. Therefore they will be clocked welll below the "maximum" potential clock speed and have as simple a design as possible to maxamize usable chips. If not Sony wont't be able to ship enough PS3's and may loose market share to Microsoft. I don't think they will risk that on to complex of a system that may limit their ability to actually deliver the systems.
Manufacturers loose BIG money on console sales. They gain it back with games. Processor cost isn't as much of an issue as it would seem.
Being that video game sales are second only (and barely) to the movie industry (and growing) for entertainment sales I don't think they are concerned about the price so much for this new generation console either. Billions, and billions.
The other thing to note is that IBM is also designing, and manufacturing the new processor for the XBOX. Sony got into bed with IBM with CELL for the Playstation first. Not that it necessarily means anything in big business anyway, it was just something I thought to add to pot, and stir.
http://www.us.playstation.com/pressreleases.aspx?id=208
There is actually quite a lot of information on Cell on the Net from patents, public talks by Sony people, and probably some people breaking their NDAs. From everything I've read so far, a Cell powered PS3 is going to be a monster performance-wise. Although it is difficult to tell which information is PS3 specific and which is referring to other future Sony hardware. It is fascinating stuff for anyone with even minimal chip tech knowledge. I can't wait to get my hands on a PS3 dev unit.
I would assume Apple has had talks with IBM for some time now about the possibility of something similar to Cell type chips in Apple hardware in the future.
Yes, the author is an idiot.
Originally posted by onlooker
Manufacturers loose BIG money on console sales.
Actually only Microsoft looses big money.
Nintendo has always been near the break even point on their console hardware. Sony hit the break even point about a year after the PS2 was released due to their ability continually decrease their manufacturing costs.
What's not to believe? This guy is crapping out of his mouth for this story.