Re: 1teraflops cell processor possible?

On Tue, 18 Nov 2003 14:29:45 -0600, "Del  Cecchi" <[EMAIL PROTECTED]>
wrote:

<snip>
>
>Don't forget the difference between "paper shows can be designed" and "is
>practical to design..."  The device has to be manufacturable (die size),
>coolable, powerable, and of wide enough interest to be worth dumping many
>millions of dollars into design and fab.  I wonder what the NRE will be at
>65 nm?
>
The Imagine processor for image processing has a similar architecture
and has already been prototyped at 15micron to produce 11.8Glop
(32bit) operating at 288MHz.  This work is cited by the SC2003 paper.
The L-cubed scaling of the SC2003 paper would put that chip at 55
Gflop on 90nm.  Not quite 128Gflop, but not bad for a start.

NRE would probably keep this all science fiction were it not for
streaming media.  Whether a version of it will happen that is
appropriate to science is another story.

RM