Re: Prometheus

I have no doubt that there could be a substantial market for
inexpensive, low CPW systems as an upgrade path for customers using
S36E or other old green-screen apps.  These systems might be a could
way to move these customers a few small steps along the modernization
path.  But I have a hard time seeing the budget model happening, for
reasons discussed below.

Perhaps a better approach is hosted services. IBM (and the BP's) can
sell xCPW worth of hosted capacity on a big box at IBM's data center,
accessible across the internet.

On 5/30/06, Trevor Perry <tperry@xxxxxxxxxxxxxxxxxxxxx> wrote:

Tom,

I suspect that if they can make a 3800CPW chip run as slow as 600CPW with
(what appears to be) little effort, then who knows how low they can go?  I
would expect, too, that the software that runs to cripple interactive down
to 30CPW could also be used to further reduce the 3800CPW to something
small. Although, it would be nice to know, I would say few customers would
need anything less than 100CPW - at least the price point should be good
enough there not to worry about having to go smaller.

Actually, the price point is kind where I am going.  The low CPW
systems have to be substantially less expensive in order to sell, and
in order to be profitable at a lower price the cost has to be a lot
lower.  If the same processor chips are used, where do the cost
reductions come from?  The answer has to be cost-reduction in the
other components (as Larry Bolhuis discussed), but I doubt that can be
done without sacrificing the high reliability of the system i.

My brain keeps saying - small i5 = portable! Then I realize the chip is the
same size no matter what the CPW restriction. Then, I look at the Xbox and
realize it has 3 Cell chips - based on Power 5. Ok, now I DO want a portable
i5!

I seem to recall that IBM produced a portable AS/400 (P02?) many years
ago.  It failed miserably then, but who knows now!