<akhan@att.net> wrote in message
news:1171373753.784135.295060@q2g2000cwa.googlegro ups.com...
> On Feb 10, 10:06 am, "JP" <piper...@swbell.net> wrote:
> > <a...@att.net> wrote in message
> >
> > news:1171034965.501078.105560@j27g2000cwj.googlegr oups.com...> I have a
client who wants to upgrade the hardware, they want to stay
> > > with Open Server 5.0.5, I am thinking about PIII machine with single
> > > processor, what is the maximum size Processor 5.0.5 supports, any
> > > specific brand name/model. I have used HP/Compaq Proliant in the past.
> >
> > > Thanks,
> >
> > > Abid
> >
> > Why stop at PIII. I've been running 5.0.5 on Pentium 4's for more than
4
> > years and have never had a problem (even on the Xeon chip). Yes, I
know
> > Xeon is supposed to have problems with any release prior to 5.0.6, but
my
> > experience has been nothing but good when using them. I have seen nary
a
> > single fried CPU yet, and I must have 15 or more clients running them
right
> > now.
> >
> > Take it for what it's worth. Action speaks louder than words.
> >
> > JP
>
> Thanks JP and Steve,
>
> I end up going to the site to check some other stuff, It turns out
> that they have 5.0.6 so now I guess I will have no problem running on
> P4 (Xeon chip).
>
> Abid
>
>
Glad you're able to move up the food chain.

Now, for all of you inquisitors who are scared to death to move to a P4
because of SCO's TA, take me at my word. I have plenty of 5.0.5 machines
running on a P4 processor; I have never experienced a single problem or
issue regarding the hardware on any of those machines; many of the machines
are in production mode 24/7 and have been running for more than 4 years. Is
this an iron-clad guarantee that they won't fail? Definitely not. But,
experience tells me that the over-heating issue with the Xeon chip and 5.0.5
is not as serious as some would have you believe.

Each of us has to use our own good judgement regarding the Xeon chip. Maybe
run a few trials or whatever. But, don't be afraid to try it just because
some TA says you 'MIGHT' experience a problem. The keyword here is 'might'.
They don't say you 'will', they say you 'could' experience a problem. It's
not a prohibition against use, it's only an advisory of something they
experienced in their testing. And as we all know, testing usually stresses
machines to the absolute limits, where-as, normal runs occupy maybe 30% of a
machines capacities.

If you have an application that absolutely crunches the machine at max
capacity, then you probably need to do some testing, but otherwise, try it,
you might like it.

JP