Sony and IBM ’s newCell processor(the one that will go into the next Playstation and future Toshiba and Sony video ) is getting a passel of well - deserved buzz at the moment . It may not stop up being the media pounding that everyone hope for , but it ’s certainly a playfulness design to ogle . Intel , of course , isplaying it down , and while they ’re likely just jealous that Sony has a pretty dress right now , the sum of their whimper is valid : wait until we can see the products .
https://gizmodo.com/ibm-and-sony-cell-round-up-32387
We linked to Jon Stokes ’ excellentdissection of the Cellon Ars Technica yesterday in our remainders ( Stokes has a gift for make something as dry as processor engineering seem not only comprehendible , but engaging ) , and in Slashdot ’s write up to the same , I found this great leaning of the ‘ Top 7 myth of the New Cell Processor ’ that capsulise a great deal of first information we have about the chip , and turtle many of the grandiloquent ideas Sony and IBM ( and Apple ) fanboys attempt to blow . I ’ll take out it after the startle .
There are so many multitude saying obtuse thing about the Cell and the forthcoming PS3 , I have to go down some things straight . Here goes :
The Cell is just a PowerPC with some spare vector processing .
Not quite . The Cell is 9 complete yet simple CPU ’s in one . Each wield its own tasks with its own memory board . Imagine 9 computer each with a really fast internet connection to the other 8 . You could problably treat them as extra vector CPU , but you ’d then miss out on a lot of possible applications . For instance , the small processors can talk to each other rather than mold with the PowerPC at all .
Sony will have to deal the PS3 at an incredible loss to make it free-enterprise .
Hardly . Sony is following the same game program as they did with their Emotion Engine in the PS2 . Everyone thought that they were losing 1 - 200 bucks per political machine at launch , but financial records have render that besides the initial R&D ( the cost of which is backbreaking to calculate out ) , they were only sell the PS2 at a little departure initially , and were split up even by the goal of the first year . By fabbing their own units , they took a immense risk of exposure , but they draw huge benefits . Their risk and payoff is roughly the same now as it was then .
Apple is going to use this central processor in their newfangled machine .
dubitable . The trouble is that though the chief central processor is PowerPC - based like current Apple chips , it is stripped down , and the Altivec sustenance will be much lower than in current G5s . Unoptomized , Apple codification would campaign like a G4 on this hardware . They would have to commit to a lot of R&D for their OS to use the extra 8 central processing unit on the chip , and redesign all their tweaked Altivec codification . It would not be a childlike port . A brace of years to complete , at least .
The parallel nature will make it impossible to program .
This is half - reliable . While it will be hard , most game system of logic will be performed on the traditional PowerPC part of the Cell , and thus normal to program . The hard part will be reduce in specific algorithm , like a cathartic engine , or sure AI . The modular nature of this code will mean that you could corrupt a cathartic railway locomotive already design to match into the 128k restriction of the subprocessor , and add the hooking into your computer code . wanton as Proto-Indo European .
The Cell will do the graphics processing , leave only rasterezation to the television card . Most likely false . The high - end video cards come out now can work the rendering chain as fast as the Cell can , look at the raw specs of 256Gflops from the Cell , as opposed to about 200GFlops from picture cards . In two years , video batting order will be capable of much more , and they are already optomized for this , where the Cell is not , so video card will perform closer to the theoretic limits .
The osmium will handle the 8 extra vector processor so the computer programmer does n’t need to .
Bwahahaha ! No way . This is a fragile spot of coding that is going to take to be tweaked by highly - pay coders for every single game . Letting on OS predictively determine what codification needs to get sent to what processor to run is harebrained in this shell . The price of shift out pedagogy is run to be very in high spirits , so any switch will call for to be cautiously considered by the couturier , or the frame - rate will polish off rock - bottom .
The Cell chip is too prominent to fab expeditiously .
This is one myth that could be correct . The Cell is huge ( comparatively ) , and given IBM ’s problem in the late past with make large , truehearted PowerPC chips , it ’s a huge gamble on the part of all party involved that they can fab enough of these things .
Top 7 myth of the New Cell Processor[Slashdot ]
Daily Newsletter
Get the best tech , science , and refinement word in your inbox daily .
newsworthiness from the future , delivered to your nowadays .
