Richard generally agrees with me, but has a different focus:

> What he said, except that I put the emphasis on lower price,
> rather than on higher performance, for the low end A class
> box.

> > I would expect that one of if not *the* worst impact of
> > running the new A-Class e3000 at ONE-FOURTH the
> > CPU speed of the equivalent HP 9000 will be in choices
> > many if not most potential *new* customers will end up
> > making;  i.e.:  ......

YPerspectiveMV, but for the "average" potential new e3000
customer I would respectfully disagree with Richard's focus;
for following reason:  I don't think it's real-world reasonable to
expect CSY to sell an A-Class e3000 for the same price (or
even pretty close to the same price) as its 9000 counterpart.
As long as the *overall* price-performance cost of an e3000 is
not in the order-of-magnitude more range, if the performance
with "standard" Internet apps is close I think most companies
will not have a problem paying a bit more for an e3000;  since
the initial purchase price of a server is at least a "once in a
while" cost; that in the big scheme of overall cost of ownership
is not nearly the biggest item.  But if it costs significantly more
*and* performance is a LOT less, that usually is not gonna fly.

Crux of the problem is that right now if you divide:

Cost of *base* A-Class e3000 ($15K, right ??)   by
Cost of *base* A-Class 9000 (about $5K, right ??)
you get about 3.   About same result for quality Win2K box.

Then if you divide:

CPU performance of A-Class 9000   by
CPU performance of A-Class e3000
you get 4.

....  Which yields an overall "*CPU* price-performance ratio"
for CPU-hungry Internet apps of about ** 12:1 ** in favor of
the HP 9000 (yes:  I realize raw CPU performance is not the
whole story.....  but it's a big part of the story).  A  2:1 or even
3:1  ratio would probably fly pretty well for many sites (and
3:1  is what we would have if the e3000 A-Class was not
crippled)....   But expect > 10:1 is a REAL hard sell to NEW
customers....

Ken Sletten