Newsgroups: comp.graphics
Path: cantaloupe.srv.cs.cmu.edu!crabapple.srv.cs.cmu.edu!fs7.ece.cmu.edu!europa.eng.gtefsd.com!darwin.sura.net!sgiblab!sgigate!odin!sgihub!zola!twilight!zola!anchor!olson
From: olson@anchor.esd.sgi.com (Dave Olson)
Subject: Re: SGI sales practices (Was: Crimson (Was: Kubota Announcement?))
Message-ID: <hdptite@zola.esd.sgi.com>
Sender: news@zola.esd.sgi.com (Net News)
Organization:  Silicon Graphics, Inc.  Mountain View, CA
References: <29542@hacgate.SCG.HAC.COM> <115072@bu.edu> <30523@hacgate.SCG.HAC.COM> <1qjrec$qem@network.ucsd.edu>
Date: Wed, 21 Apr 93 05:14:50 GMT
Lines: 90

In <1qjrec$qem@network.ucsd.edu> spl@ivem.ucsd.edu (Steve Lamont) writes:
| What I *am* annoyed about is the fact that we were led to believe that
| we *would* be able to upgrade to a multiprocessor version of the
| Crimson without the assistance of a fork lift truck.

It should have been made fairly clear that the *most* Crimson would
ever get was a 150 (75 old style) MHz CPU upgrade.  Certainly this
was mentioned on comp.sys.sgi on more than one occasion as being
likely.  If our sales folks were saying otherwise, they were either
confused, or less than honest/ethical, or somebody further up the
chain inside SGI was misleading them.

| I'm also annoyed about being sold *several* Personal IRISes at a
| previous site on the understanding *that* architecture would be around
| for a while, rather than being flushed.

There were 4 versions (20, 25, 30, 35), although admittedly the 30 came
out at the same time as the 35, over a period of 2 1/2 years.  The
chassis simply couldn't be pushed any further.  I'd say 4 years was
a pretty good lifespan, myself, for a system design in this day and
age.  Getting the 35 to work caused a lot of gray hairs in both the
hardware and product design groups; we would have been out of our
minds to push it further, and I *know* that was made clear, almost
from the day the 35 started shipping.  We had one last kicker in
the form of the Elan graphics, which made 3 graphics versions over
its lifespan, which I also think is pretty good.

| Now I understand that SGI is responsible to its investors and has to
| keep showing a positive quarterly bottom line (odd that I found myself
| pressured on at least two occasions to get the business on the books
| just before the end of the quarter), but I'm just a little tired of
| getting boned in the process.

Please, by all means send a complaint letter through SGI support
or sales on your concerns.  There should be no reason for sales folks
to misrepresent future upgrades to customers (sure, sometimes there
will be confusion for a while, over whether an upgrade will be available,
but that shouldn't last too long, and doesn't seem to be what you
are referring to).

Yes, the sales folks *do* get bonus's at the end of some (all?) quarters,
but that is pretty common industry wide, and sometimes that can result
in good deals for customers (sometimes it probably pushes folks into
systems that aren't what they need, I'm sure, but nobody is *forcing*
you to buy at end of quarter, after all...)

| Maybe it's because my lab buys SGIs in onesies and twosies, so we
| aren't entitled to a "peek under the covers" as the Big Kids (NASA,
| for instance) are.  This lab, and I suspect that a lot of other labs

They don't get all that long a lead time either; although certainly
they get presentations on possible new products, and their opinions
may well influence the end product, but that also is life in the
industry.  We can't design systems that meet just their needs, or we
won't sell too many systems, after all (which is not to say that we
don't have some niche products, like Reality Engine).

| and organizations, doesn't have a load of money to spend on computers
| every year, so we can't be out buying new systems on a regular basis.
| The boxes that we buy now will have to last us pretty much through the
| entire grant period of five years and, in some case, beyond.  That
| means that I need to buy the best piece of equipment that I can when I
| have the money, not some product that was built, to paraphrase one
| previous poster's words, 'to fill a niche' to compete with some other
| vendor.  I'm going to be looking at this box for the next five years.
| And every time I look at it, I'm going to think about SGI and how I
| could have better spent my money (actually *your* money, since we're
| supported almost entirely by Federal tax dollars).

But surely you don't expect a system you buy now for a five year
period to be constantly upgradable over that entire five year
period?  That's a rather unreasonable expectation, in my experience
(with workstations/microcomputers).  Supported, and parts available,
yes, but certainly not upgradable to the latest and greatest!

| Now you'll have to pardon me while I go off and hiss and fume in a
| corner somewhere and think dark, libelous thoughts.

I missed your first posting, but as I say, by all means share your
frustation with somebody at a level inside SGI where it might
have an effect (not immediate, I'm sure, but complaints aren't
going to be ignored, and *may* affect future plans, if we 
hear similar things from more than one person/site).

All of the above is, as usual, my personal opinion, not SGI's.
--
Let no one tell me that silence gives consent,  |   Dave Olson
because whoever is silent dissents.             |   Silicon Graphics, Inc.
    Maria Isabel Barreno                        |   olson@sgi.com
PS: I start my sabbatical 29 May, ask those questions now ;)
