Cell processor death

Question

In recent times, I have heard many people claiming that the Cell processor is dead, mainly because of the following reasons:

• Lack of support in the new playstation 3, because the user can not install linux
• GPU increasing computing power and cost reduction
• Existence of a unified programming approach (openCL) for different GPUs, and not for CBE (today it was announced for Cell!)
• Karenia real cell use examples (besides academia)
• Global feeling of failure

What do you think? If you started programming a cell two or three years ago, will you continue to do this or are you planning to switch to a GPU? Is there a new version of the cell?

thanks

Answer 1

I would say that the reasons for the lack of popularity for cell development are closer to:

• Lack of success on the PS3 (due to many bugs from Sony and intense competition from the XBOX 360)
• Low production income, high cost (partly due to low yields) and the lack of affordable equipment systems other than PS3
• Development complexity (the cell is an unusual processor for design and the tools are not enough)
• Inability to achieve significant differences in performance compared to existing x86-based hardware. Even the XBOX 360 with its multi-year triple-core Power Architecture processor has proven its competitiveness compared to the modern Core2 Quad processor, the advantages of the cells are not obvious.
• GPU, CUDA

Answer 2

1000- , 10- . 1000 , . , , . Cell , 10s, . ( , , 100 .)

IMHO Cell . ( Cell MICRO 2005 . , .) - GPU , . . IF, , . , .

, - , , D.

, parallelism , Cell, . Cell , , , , . .

, , $ D $. .

Answer 3

?

1) SDK . , , IBM, Cell SDK.

2) 32 .

3) OpenCl 3-4 .

Answer 4

, GPU?

I would think that 90% of the people who program the Cell processor are not in a position where they can arbitrarily decide to stop programming for him. Are you asking this question in a very specific developer community?