At 10:12 10.09.98 +0100, Roy wrote:
>You have to be looking at how you can bring people to the HP3000. Which
>means they have to be able to compare an HP3000 solution against any
>other on 'bang per buck'.
Good point. But I wonder if a TPC number would be very helpful in this
regard. I am not familiar with TPC benchmarks, but as a customer I would
probably not be running TPC programs but real applications and thus some
info like this-or-that hardware with this-or-that application can make
this-or-that amount of users go along doing productive work...
I always found (real world) examples like 30 users happily running MM/PM
on a 3000 series 70 with (forgot the MHz) and 8 MB or main memory much
more impressive and plastic(?) than x tpC or y specint. Especially when
comparing to a typical piece of desktop hardware today... (or what those
SAP shops might be throwing in hardware-wise).
The non-tpC examples also take into account that many HP 3000 applications
seems to be written in a much more efficient manner that typical HP 9000
or Client/Server apps (which feel like "through anough hardware and/or
consulting at it and one day it will finally work").
Lars (only speaking for lappel, as ususal)
PS: I don't recall if the 30 users example above was the real number...