Conference 7.286::digital

    >TRUE OR WHAT ?????
	
    	What...

     Not true. Would you dare to publish here the source of this silly
    rumour (like the Mitsubishi/KO retiring rumours). ?
    
    Plans are in place to manage the VAX9000 situation on an account basis
    in a professional, unhysterical, positive manner as befitting the
    "best" computer company in the world.
    
    Ian
    

    Re 1632.2 - Would you like to publish what these plans are,and also to
    say what your sources are!!
    
    I am currently handing several VAX 9000-410 accounts who have ordered
    but not had delivered, their 9000 based systems. If they hear that 
     9000's are to cease production, and also that 6000-600's are benchmarked 
    as being faster than 9410's, they they will cancel in favour of the new
    machines, with a great loss of profit to the corporation.
    
    What is going on ?? - what happened at the Exec meetings of 8th Oct??
    Are we still in this high-end market or not??
    
    
    Confused
    
    
    
    
    

    re .1....agree that this is a dangerous rumour, but your comments is
    interesting.  If we are the "best" computer company in the world, why
    would we not go to your 9000 customers, hat-in-hand, and explain
    exactly what is happening to 9000's (assuming anyone knows he said
    while cackling hysterically)??  I can't see how the impact of their
    ultimately finding our anyway could be anything but VERY negative in
    the long run.
    
    So we lose some short-term bucks, so what??  I think we'd win a
    customer-for-life is we were totally open about the whole thing.  I
    think there is a guy named KO that always talks about doing the right
    thing.  This situation certainly seems to fit that comment well.

    
    
    	OK so what *is* the truth ?
    
    	

>                     <<< Note 1632.3 by BAHTAT::SCOTTJ >>>
>                       -< Is there anybody out there? >-
>
>                                                    If they hear that 
>    9000's are to cease production, and also that 6000-600's are benchmarked 
>   as being faster than 9410's, they they will cancel in favour of the new
>   machines, with a great loss of profit to the corporation.
    
    I'm curious.  Is this true, and if so, based on WHICH benchmarks? 
    Specmarks? TPC-A? I can see a 600 showing better CPU performance, but I
    have my doubts about combined throughput.  What is the source of this
    information?
    
    tim
    

    
    
    	Check out VMSMKT::VMS_PARTNERS note 552.39 for a heads
      up to the field on the VAX 9000.  I don't know if this is 
      a members only conference or not.

It's members only.

    If the rumor about the 9000 was true, I don't think they would have
    herded all of us into the MET cafetaria this week to tell us about
    the next version of the 9000 that will be announced at the end of the
    month.
    
    B.

    Here in New Jersey we have heard unconfirmed grape vine that the new
    6600-xxx will support up to 8 processors, have expander cabs, and
    totally support the new "9000" internal controller and devices. It will
    be touted as the "poor man's 9000" and virtually the same or better
    performance. I don't see this as a unfounded rumor as we "DEC" have
    shot ourselves in the foot before, so why would this be any different??
    I don't have to remind anyone of the debacle of the 8000 series do I??
    	regards
    	John

    To the base note... rather then starting rumors and or misleading the 
    people reading this note why not ask your plant manager about it. He
    was involved with the decision. If he doesn't give you the answer, then
    stop by my office and talk to me. I was also involved and would be
    willing to have an informed conversation on the direction of the VAX
    9000 and it's impact on your group.
    
    This topic is too big to be spreading rumors....it could destroy what
    little is left of our stock price. 

    Re .6 
    
    A marketing source - usually reliable, saya that the 30th Oct
    announcements will show the 6000-610 as producing 80 tps using TPC-A
    with ACMS and Rdb. The comparable figure for VAX 9000-210 was 68.4 tps,
    and about 38 for the 6000-510.
    
    If these figures are true, then we've got a major credibility issue
    with the World - we always sold the 9000 on its strong I/O capability.
    How do we explain to customers that a 100MB/sec XMI bus on a 6000 is 
    really faster than a 2GB/sec crossbar switch?
    
    Regards
    
    John
    

    
    	re .13
    
    	Remember that the 2GB/sec crossbar switch gets fed by one or more
    	100 MB/sec (false and misleading number, btw) XMI's.
    
    	John

    
    
    	The 9000 will not be cancelled.  They will merely do "made to
    order" vs building up an inventory.
    

r.e .previous    
>    If these figures are true, then we've got a major credibility issue
>    with the World - we always sold the 9000 on its strong I/O capability.
>    How do we explain to customers that a 100MB/sec XMI bus on a 6000 is 
>    really faster than a 2GB/sec crossbar switch?
    
    I havent heard the rumours, so I wont comment on them.
    
    I dont see what the problem is. The 9000 has strong I/O capability.
    You cant deny it, it is true yesterday, today and tomorrow. There seems 
    to be some sort of revisionist mentality that creeps in every time we 
    release a new system. We start to hide from our own abilities to
    produce good products and act sheepishly. In my opinion its the 
    down faces and the mumbling that will do more damage than a new
    product release. Customers believe you have hurt them when you 
    behave like you have done so. The 9000 competes well. Its not perfect,
    so what, it does a lot very well. Of course its going to be surpassed.
    In the past, we have gone for leading edge approaches to give customers
    the performance they want. Sometimes the opportunity pans out over a
    longer term, sometimes not. Sometimes the high end is riskier, with
    more dead_ends and unknowns. A more straightforward approach may turn
    out to be a better path. I see many people running for cover over the 
    9000, these people are stupid. In any solid investment portfolio, you
    have a mix of low, medium and high risk ventures. Often the medium risk
    pans out to better returns over time. Often however, the lessons learnt
    from the volatile investment will show in better decisions down the
    track.
    
    I realise that 9000 customers will question ANY new release that
    potentially could devalue their systems, if it doesnt happen sooner, it
    may happen just a little later, but it will happen. And it will happen
    to the 6000, and the 4000, and anything after that. We are not in a
    world where you can afford the luxury of not obsoleting your own
    systems, whether with a follow-on 9K, a faster 6K, a RISC from either
    MIPSCO or DEC. If you dont obsolete your own systems, who will? I'd
    rather it be us. Customers have purchased against current technology,
    they could wait for tomorrow, but its always a day away, whether they
    are buying mainframes, pcs or walkmans. 
    
    regards,
    John

    re .16
    
    Finally, some good sense.

At the request of a noter, I have changed the title of this topic to something
more descriptive.

Bob - co-moderator DIGITAL

    
    	There is no doubt that Digital has done a very poor job in
    	bringing the 9000 to market and we still do a lousy job of
    	positioning it against the 6000 series.
    
    	I have been heavily involved with the 9000 almost since it
    	was first introduced....the team of people I have been associ-
    	ated with over that time in EIS has been responsible for doing
    	a lot of work through benchmarking and system optimization 
    	projects with customers to maximize the performance of 9000
    	and other large VAX systems in production systems applications.
    
    	In particular my group worked with Burlington last spring on
    	the Contel benchmark on a 9420 which I believe is the largest
    	RDB benchmark that has been done in Digital.  We recently took
    	did a quick look at the 6600 to get an idea of how we might
    	have done that same test with a 6600 and concluded that there
    	wasn't even enough XMI ports on the 6600 to connect up all
    	the I/O we used in that test.  And the system we tested for
    	Contel on a single 9420 was a subset of the final implementation
    	which would have required a total of 1t least 3-4 9420 systems.
    
    	We are currently working with a number of customers who have
    	very large commercial production applications running on IBM
    	3090's using the SAS application.  Some of these customers are
    	taking up as much as 30% of a 3090-600J for that application
    	alone.
    
    	There is no question that you need a machine the size of the
    	9000 to be a serious player in that market...unfortunately,
    	it is a market that Digital does not have a lot of experience
    	with and doesn't understand very well.
    
    	The 9000 is an excellent machine for large commercial applications.
    	I believe that the relatively weak success of the 9000 is not
    	due to the machine itself (although some of the early reliability
    	problems didn't help).  I believe that we have failed in two
    	major areas:
    
    	   1.  Marketing/Sales  The marketing message associated with
     	          the product was extremely confused...is it just a
    	          big VAX or is it a mainframe?  and we tried to sell
                  it the same way we've sold all other DEC systems when
    	          a real solution-selling approach is needed to penetrate
    	          many of the markets its sold into (especially IBM)
    
    	   2.  Systems Engineering  We have done a poor job of building
                  a good capability to provide the sophisticated systems
    	          engineering skills for a machine of this complexity.
    	          In our usual fashion, we expect our customers to figure
    	          out how to make it work effectively with their application.
                                                                 
    		  We continue to do a very poor job of providing the systems
                  engineering to help customers understand why this is a
    	          better machine for a lot of large applications than a
    	          6000....it isn't just a simple matter of comparing 9000
    	          VUP's to 6000 VUP's.
    
    	I for one believe that there is a very good market for the 9000
    	and I will be very disappointed if we abandon that market.  It
    	has a lot higher profit potential than many of the other products
    	we have can possibly deliver.  I do think the window of opportunity
    	is closing; however, if we can't turn this situation around very
    	soon.
    
    	That's my opinion based on a lot of experience with the 9000 over
    	the past two years....I am a little disturbed by some of the comments
    	I see in this conference about the 9000...many of the things I see
    	seem to be based on little information and could be potentially
    	very harmful to Digital's success.
    
    
    	Chuck Cobb
    		                                                             
    

    ...and another big mistake is that we also attempted to service 9000's
    in MicroVAX II mode.......something completely unsatisfactory in the
    mainframe world.
    
    ...I dunno, "build-to-order" sounds an awful lot like an obit to me.

    Large production systems are very much a different anamial than what
    most of us are use to.  I was the system manager/engineer for the
    Contel benchmark mentioned in 1632.19 and I don't think the 6XXX
    product as we know it today would be able to handle the production
    environment.
    
    I have been following this topic from the technical side and have the
    full confidence in the 9000 as a solution.  The same rumers have have
    been occuring on other notes as we've all been waiting for the end of
    October to hear the official announcements.  It's not the 6000 that is
    doing in the 9000 but it is us and our internal rumers which get out to
    the field.  Included is my reply in the VAX_9000 conference.
    
    Lee Scott
    
              <<< BTOVT::ALTSYS:[NOTES$LIBRARY]VAX_9000.NOTE;1 >>>
                      -< VAX 9000: Product Of The Year! >-
================================================================================
Note 436.12                   "End-of-life" phase?                      12 of 13
SOJU::SCOTT "Looking towards the sun"                37 lines  11-OCT-1991 16:02
                          -< Don't kill a good thing >-
--------------------------------------------------------------------------------
    I have been around the 9000s and production systems for a while
    and I've heard a number of stories about the 9000 being dead for about
    a year now.

    I personally don't believe the 9000 is dead yet due to the limitations
    of not only the 6000 but the XMI bus itself.  The XMI bus has only 14
    slots and Large production systems require large amounts of high speed
    I/O and that equates to many I/O interface units.  Typically we are
    talking about 4 CI paths and maybe more along with 3 to 6 KDM70s with
    ESE20s for high speed.  Couple that with the need for maximum memory
    and 2 to 4 processors and 2 to 4 vector co-processors - well, what XMI
    based system will come close?  There are other technical reasons why
    the 9000 class machine is better suited to the production systems.

    I think the main reason we are hearing about the demise of the 9000 is
    due to our inability to sell and service production data centers and
    therefore the 9000 is ahead of it's time for DIGITAL.  We are working
    on the next generation of cpus in which we will base our company's
    reputation on much in the same way we do (did) with the VAX 32 bit
    machine.  Unless we learn how to meet the needs of large production
    data centers I'm afraid that no mater what we come up with in the
    future to replace the 9000, the same thing will happen and we will be
    limited to selling workstations to universitys and scientific labs.
    Is that to be our lot in life?

    The 9000 or the need for a 9000 class machine is not dead but
    speculation about it from within DIGITAL and passing that speculation
    on to the customers will surely kill it faster than the next generation
    machine coming to market.  The 9000, if supported by us will have
    at least 2 or 3 more years of life in the market place and will be
    around for at least the next 10 years.  If we don't learn this now, then
    I'm afraid the 9000's replacement will have even a shorter life cycle.

    Don't let me be the only one touting the 9000 and it's capability.

    Lee
    
    

    My brother is a product support manager for an IBM compatable mainframe
    company.  He has told me that the competive analysis reports he has
    read in the industry speak well of the 9000.  
    
    If it dies, it is only because we are not a mainframe class company. 
    The 9000 surely is.
    
    It may have had some problems with the first ones shipped.  The first
    mainframe my brother was responsible for supporting had a MTBF measured
    in minutes.  Something as complex as a mainframe is going to have initial
    problems.  Thats life for us, IBM, Amdahl, Tandem and all the rest.
    Customers initially expect problems.  They also expect outstanding,
    immediate support.  Can we do that??  Do we do that?
    
    We should stay the course with the 9000.  It will sell.  The selling
    cycle may be two years instead of two months, but it will sell, if we
    have the application software that a customer needs in a production
    environment.
    
    It is not the 9000 we should be questioning, but ourselves.

    re .-1
    
    Exactly. This company seems to have largely lost the ability to work
    sales cycles longer than about 2 months. Yes I see some sales execs
    being very successful with long sales cycles but not many. In order to
    sell 9000s we've got to learn how to go for that long term large
    profit. That means sales managers have to learn how to evaluate sales
    people using some other performance metric than how much revenue/profit
    did you bring in this quarter.
    
    I'm in IBM Interconnect engineering and I'm amazed at how easy it is to
    influence customers on the benefits of our product set when I get to
    talk to customers. It needs the technical side of the sale. So what is
    our wonderful field organization doing right now. Yes you've guessed it
    laying off very knowledgable network sales support people because the
    AGMs can't work out how to pay for them. I bet 9000 sales are very
    similar to how IBM Interconnect has to be sold. What nobody seems to
    realize is it is damn hard to get into a new IBM account. But once
    you're in there is a lot of revenue/profit to be made. The 9000 is a
    GREAT way to get into IBM accounts. But it just won't work with our
    current, flog 'em a cheap hot box at a loss so I can make my numbers.
    
    Disgusted of LKG

    Perhaps one of the big reasons that the 9000 is not selling is the
    goaling system under which our sales people must operate.  A long term
    opportunity is one which we expect to close in the next quarter.  Now
    with the TFSO 4 just completed, there is an absence of sales support
    people which makes the problem more acute.  Sales people are especially
    scrutinized now for near-term results, they must sell something now or
    they may be the on the next TFSO 5 list.
    
    They are not bad people, quite the contrary, but they are under
    tremendous pressure to produce.  Our sales people are rotated from
    account to account very often as opposed to IBM where they sales types
    die or retire before they are replaced.  A sales type is not going to
    start a major sales effort that will not close for two years in Q2 FY92
    because he/she will almost never see the fruits of their efforts.  Why
    do something to save the next guy's job when I may loose mine in the
    process.
    
    For any large CPU to be successful requires a major effort to prospect
    for new applications in an existing account or replace a competitive
    system in a new account.  New accounts are covered by General Business
    Account Reps who work with OEMs or distributors.  They must really
    scramble, they can't afford to prospect.  They are measured for sales now
    so they spend their time for short term opportunities.  Also the
    General Business Reps are usually the newer sales people so they are
    less experienced selling large systems like the 9000.
    
    Selling the 9000 takes a long time to identify the opportunity and
    there is going to be a lot of competition, especially from Big Blue.
    It will take a team of people and it will take benchmarks and more
    benchmarks because IBM will protest the success that we might enjoy if
    we win.  They are masters at protecting their installed base because if
    they lose an installed customer, they are probably history from IBM.
    
    I submit that selling any large system like the 9000 will require a
    different type of measurement goals which is not going to happen with
    the situation that the company is presently in.  It will also take
    senior sales people and good sales support people who can spend time to
    identify the opportunity and work that opportunity until it closes. 
    Until that happens we can design the best product available and it can
    be acclaimed as the Product of the Year, but the results may be less
    than gratifying.
    
    Regards,
    
    Bill
    experienced selling large systems like the 9000.

    Wow, I can't believe some of the off-base rumors that I've seen here,
    they don't help us, and they must stop.  The field will be learning
    through a DVN and in-field training this Thursday what is going on with
    a lot more than just VAX 9000, and it's very good news.  Yes, there is
    an announcement coming soon, and questions regarding the entire VAX
    family will hopefully be answered.  I suspect that it would not be
    appropriate to discuss unannounced products or directions in this
    forum, however to help us readjust our thinking:
    
    	- Do not confuse underlying H/W technology changes with changes to
    the VAX/VMS computing environment.  Over the years we have constantly
    changed technology, platforms, and H/W implementations without severely
    compromising the industry's best EVER story for investment protection of
    S/W, people, and even usually storage and networks as VAX systems
    evolved.  The VAX/VMS computing environment will not only continue,
    spanning desktop to mainframe, but will regain new and continuing leadership
    performance and price/performance.
    
    	- The IRRELEVANT question to ask:  What is the status/future of the 
    VAX xxxx cpu/cabinet/platform?
    
    	- The RIGHT question to ask:  What is the status/future of the
    VAX/VMS computing environment, why will it keep getting better, why
    should I buy now, and how will you protect my current investment?
    
    	- The VAX/VMS family of systems will continue, and keep getting better.
    	
    Cheers,
    Ken
    

    I haven't seen an effective rebuttal to the widely circulated Gartner
    Group analysis that the depreciation on a $ per VUP basis for the VAX
    9000 over the next N years, calls into the question the rationale for a
    VMS customer buying it, or for DEC to sell it.  Maybe VAX/VMS lived,
    lives, and will live on other platforms but the VAX 9000 is what's in
    the price book to sell now.
    
    Maybe DEC should continue to pour those dollars into making bigger and
    better VAX's for those contented VAX customers, even if we ultimately
    lose money on it.
    
    In areas when the computer marketplace is actually growing, I'd like to
    see our products making the other guy's contented customers less so,
    and enlarging our customer base in a profitable way.
                                                        

Does anybody think the lackluster sales of all of our products might just 
possibly be related to the lack of money on the part of our customers?  Or 
is it lousy management on our part?  EVERY computer company's sales are 
off!  Who's fault is that?  Is it a normal economic cycle?  

I contend that it doesn't matter - how a company manages itself through
these kinds of times will dictate whether or not it survives.  It has been
called survival of the fittest!  The VAX9000 is a solid computer with solid 
engineering behind it AND a solid company to back it!

Bruce

    
    	I applaud those responses that are telling the entire story of the
    	VAX9000. I think that .19 said it very well.
    
    	I too was involved with the VAX9000, in the area of Customer Svcs
    	training. When you look back, we tried to do a heck of a lot with
    	the 9K. The technology alone was incredible...an MCU is not some-
    	thing that you can just slap in and expect it to run.
    
    	Some things we succeeded with, some things we didn't do so well on.
    	But it wasn't the machine itself...the economy had a lot to do with
    	it, too.
    
    	I don't know too much about how our sales force works, but from
    	what I hear, IBM sales people spend a lot more time with their
    	customers before AND after the sale making sure that they are
    	happy. I'm not quite sure that this is how DEC works...and if
    	you are trying to break into a $30-$40 billion mainframe market,
    	you have to be quite sure that your image as a mainframe vendor
    	will certainly be under the magnifying glass from the industry.
    
        re .20
    	> ...we also attempted to service 9000's in MicroVax II mode...
    
    	This is definitely not true from the standpoint of the Field 
    	Service troops..it may be true from other angles of the company.
    
    	It would be a shame to see the 9000 retired so early on, but I'm
    	afraid that those decisions, whether they favor the 9000 or not,
    	have already been made, and nothing we say in this conference is
    	going to change it one way or another.
    
    	John

RE: .28

> EVERY computer company's sales are off!  

Not true at all.  The computer industry as a whole is doing quite well
in the recession.  It's the high overhead companies that are doing poorly.

Where'd the base note go??? 


The original query in the base note asked: "Where are they now?"  
I will agree that I was also uncertain as to what was being asked.  
There seemed to be a reference to Digital's penchant for positioning
products against one another.  I remember a rather uncomfortable
afternoon in a customer's office listening to him shout about how
we hurt his company financially and his prestiege personally by the
quick follow-on release of the 8000 series products.

It could also have been a reference to one of several different 
aspects of the VAX 9000, for example:

	Where are the orders?
	Where are the installations?
	Where are the sites/customers?
	Where are the demos?
	Where are the trained people to work on it?
	Where's the advertising?
	Where's the marketing?

Back in 1980 when there were hundreds of orders for VAX-780s, there was 
considerable commitment to training from customer service.  I don't believe
we have that same whole-hearted enthusiasm to support the VAX-9000 we had
then.  Granted the product is vastly different; the user environment is not
very friendly; and the market is dominated by a powerful adversary; still
we must face the fact that we squandered huge sums of money for poor ROI
and indentured our revenue-generating groups like customer services to a 
frantic no-win battle of diminishing returns.

Customer service is forced to economize on facilities, workspace, carplan,
travel, and training; in fact, all aspects of expenses which show no 
tangible and immediate impact to the bottom line (the profit benchmark).
For example, when a manager wishes to improve his balance sheet immediately 
so he can "look good" and make his numbers, he cancels all training out of
the district for the next six months!  That has happened in some cost
centers.  And this style of management by convenience at the local level
has a familiar ring at the higher levels as well (TFSOx).  If the shoe
doesn't fit, lop off some toes.

I'm not saying I know the answer or even have a clue to restoring Digital
to profitability, but I know this: you don't compete in this mainframe
marketplace by taking a hide and watch attitude.  Test marketing is over,
I think.  I just can't help feeling we went into this 9000 project as an
experiment.  I do not see commitment.  Maybe we couldn't afford it.  Maybe
this was the best we could do.  Maybe we're in too many markets, with too
many products.

I do not know how to demonstrate to a customer that we deserve his trust
and his business.  On the other hand, I can question my company's sense
of judgement when they think they can get by with less than a total
effort to support a sophisticated product in a sophisticated environment
with the same old "business-as-usual" onesey-twosey mentality with regard
to training, logistics, and capital equipment.  I fail to see how you can
motivate people to share your vision of "customer service" when daily
they see the countless incidents and examples of missed response times,
no spares or parts, DOA parts, calls dropped, poor coordination and
call escalation, extended outages due to the "warm body" syndrome.
I left the field partly because I got fed up with the hypocrisy and
lip-service we pay to the slogan "The Light Is On The Customer".  I'll
tell you, if the light is on the customer, it must be infra-red, because
I don't see it.  Sure, there's bright spots here and there, but only
when an individual account rep or manager personally commits to a work-
ing repore with his customer.  And today that's rare.

There's another element to the situation that points out a serious flaw
to a service plan on a complex product.  Time spent in training.  In the
case of the 9000, the complete training program requires months.  This
program and this training opportunity is not assigned indiscriminately.
Only a very few are picked; usually a support engineer,  and the account
rep gets the abbreviated crash course of two weeks.  That is the way it
used to be and I doubt it changed much.

Training on new state-of-the-art equipment has always caused friction
and ill feelings in the ranks of the field service pecking order.
Far too frequently training is bestowed as a reward or granted for the
wrong reasons; i.e. training the wrong people.
Another not so confidence-inspiring sight for a customer is to have an
apprentice CSE receiving OJT on site on a live system.  But, again, this
also is a result of not having enough money for training in the budget.

Digital is having a very difficult time in this new user/client market.
We talk a good game, but it loses something in the translation.  We have
impressive new products, spiffy new chips, but we don't want the customer
to burden us with comitment; partnership.  We want him to buy our quite
expensive products.  We want him to be not merely impressed; he must be
in awe.  But don't ask us to provide solutions or be a business partner.
We have enough entrepeneurial spirit within Digital to give us a major
migraine; what with all the tail wagging the dog.

We're being slowly nudged out of every niche we ever hoped to occupy.
And unfortunately, we don't have the financial clout or wherewithal
to help us nudge back in somewhere.  Time is definitely running out on us.
In fact it may already be too late.  I see little to be optimistic about.


imho,
Mike

    Mike's .31 note highlights a point about committment that needs to be
    stressed. Confidence is not something that a customer gets from one
    enthusiastic person, they have to see it in every interaction they make
    with a company. When an Engineer commits to deliver, he should make
    that committment on the full belief that the company has the resources
    to back him. When we say YES, that has to be in the knowledge that
    because you have said YES, the person above should go with it because
    the committment is to the customer.
    
    When we fail to put the logistics in place to service the customer
    properly, the training, the money to sell the product in the minds of
    the customer, we send a message. We tell them the confidence they see
    in the man in front of them is not backed up by the people who provide
    the resources that he is promising.
    
    Look at the Toyota LEXUS. A perfect example of a companies first foray
    into a new market. A market characterised by high prices, complex
    technology, a requirement for a lot of support and handholding, a 
    niche where image is all important. Sound familiar? No doubt Toyota
    has made mistakes entering this market, but at any time does anyone
    doubt their committment to 
    	a. get the engineering right
    	b. display the confidence they have
    	c. impart that confidence to the customer
    	c. build the image
    	d. attempt to do all of the above a little BETTER
    	   than the incumbants because they are the new entrant.
    
    Initially, they didnt get off the ground too fast, sound familiar?
    But did anyone hear them waiver in their resolve? 
    
    Now, we could learn a lot from companies that have gate crashed new 
    markets with similar characteristics to our own. When Toyota delivered
    the Lexus, they realised that in that market segment you dont just
    deliver a car. In fact, the car is only about a quarter of the package.
    They produced new showrooms, specially trained mechanics, special
    levels of service handling designed to SURPASS their opposition. 
    You know they are serious and to Toyota, this car is special. If it is
    to them, then maybe that feeling of regard is passed on as well.
    
    With the 9000, we bumped service up a bit. But when I go to parties
    frequented by computer nerds from the IBM world, its like we are on
    a different planet. They have no perception of us. You are already
    behind the line because this is where the confidence building comes in.
    You have to have them talking about the new kid in town who makes a 
    splash. If we are not prepared to wear our pride on our sleeves, and
    CONTINUALLY drive the point that we are better AND back that with
    service and confidence. Who is going to believe us when we say we are
    best? 
    
    Let's say it all together, 
    1. the 9000 is an excellent mainframe system. 
    Now try and say 
    2. we deliver excellent service, 
    3. have excellent logistics in place to service it, 
    4. the excellent market presence.
    5. Now say we have full committment from everyone.
    
    Oh well, 1 out of 5 isnt bad. The problem is your competition's
    systems are good too (more costly for the same level of excellence),
    but they play the game a little better in these other areas.
     
    If Engineering produces excellent Alpha systems, the customer may give
    us 1.5 on a tick chart. We need to do a lot of LEXUS pre-work and 
    make sure that all the ticks are in place and to deliver a committment
    from all levels that will achieve a level of service above our
    competitors.
    

    Re .30
    > The computer industry as a whole is doing quite well in the recession
    
    I think I must be living on a different planet to you! ;^)
    Microsoft & Intel are doing well, but MOST other companies seem to be
    having a difficult time.  From VOGON News over the last five days:
    
    IBM 3rd � profits down 85%; now eliminating 20000 staff (up from 17000)
    Sun stock fell 9% amid falling earnings expectations
    MIPSCO 3rd � losses widened from what company expected
    Borland acquires Ashton-Tate
    Floating Point Systems files for Chapter 11 (bankruptcy law)
    Motorola 3rd � net dropped 9%
    IBM France cuts employees by 10%
    
    Wang?  Unisys?  Data General?  Honeywell?

    
    
    re .31 & .32
    
    I don't know about the areas that you are comming from, but in my
    area(North Central States Region) well before a 9000 is shipped the
    account rep and backup rep are identified and trained on the UPC(1
    week) and VAX 9000 fault analasys(3 weeks).  In the case of my machine
    I was sent to the Burlington plant for the final acceptance and
    tear-down(kind of like in the LCG glory years).  The support that we
    have received from Logistics has been flawless so far.  As far as who
    gets trained, the prerequisits for the class dictate that only a highly
    VMS and VAX architecture type person should be sent.  These are usually
    senior type engineers.  Our ratio of trained engineers to machines is
    1.4 to 1 for the district.  I am in class with an engineer from the
    Florida District their ratio is 1.25 to 1.  
    
    The biggest problem I can see with the product is all of the doomsayers
    trying to kill it.  
    
    I know you might be saying that we are trying to kill it with the
    6000's, but no 6000 can compete with a 9000 in the market for witch it
    was designed.
    
    Just the way I see it
    
    Jon Aller
    Customer Services Engineer responsible for 3 9210's

    I'm sick and tired of all the people that are trying to force the 9000
    to fail. I'm so glad to see that some people that know what they are
    talking about are rebutting some of the skuttlebut.
    
    Why can't people realize that you can't bust your way into the
    mainframe market overnight? You need dedicated people over the long
    term. I believe that these people exist within DEC. Let's try to help
    them rather than trip them up.
    
    I'm the first one to want to see information openly disseminated. But
    let's get all the facts (or admit what is missing) before trashing a
    perfectly good program.
    
    Dave

    I was just on a conf. call with a customer on the West Coast who
    purchase a cluster of 2 9210's.  They run a 24-hour production system
    and cannot afford any substantial downtime.  During the past 6 months,
    they have had over 20 episodes of unscheduled down-time.  The worst of
    it is that there have been periods where it took over 8 hours for the
    appropriate parts to be located, delivered and installed!  Needless to
    say the customer is P*SSED OFF, and the local SWS guy I deal with is
    *real* unhappy.

    
     We are at a government site running a 24-hour production system with
    a 9410 and the thing has had 1 episode of unscheduled down-time due to
    THE CUSTOMER. Our hardware service record is impecable. Those guys
    shine. The customer is real happy, REAL HAPPY. The local on-site SWS
    guys (myself included) are at ease. So what is the real story here?
    
     Oh, this 9000 has been around for six months also.
    
    Mike Z.

    Interesting that we program announced a new 9000 on the DVN this
    morning...
    
    New Vax 9000
    
    - faster performance (would we expect Slower?)
    - Upgradable to RISC VMS
    - Available in Summer
    - Upgrade program from today's VAX 9000
    
    (slide #41)
    
    

    re -1.
    
    What it means is that the VAX9000 as we know it will be alot different
    in the summer.  The platform as people will find out soon will be
    different than the current platform.  I will not comment on all the 
    various rumors that we in BTO have heard, but I will say that the
    follow on VAX9000 will be coming out of the mid-range group.  As for
    downtime at the customer site, we have heard of a couple of cases; some
    of this down time was due to the customer.  Yes the 9000 has its
    problems but it is a good machine.
    
    -Harvey

re: .39

	We saw this smoke campaign in memos from a certain VP
this summer when the real follow-on VAX 9000 work was canceled and a lot
of the engineers were told to find work elsewhere.

I won't comment on what this thing really is, but, it ain't no "VAX 9000".
For me this was the final nail in the coffin of the real VAX 9000.

    The customer site I work at has three 9000's, and has had very little
    problems with them.  They do however have a LARGE problem in the
    overall support for the mainframe environment, or lack there of.  This
    site has 6 large clusters (minimum of 7 cpu's/cluster), and enough disk
    drives to fill a warehouse.  The problem is we have the hardware, but
    not the tools to manage a large data center such as this.  The
    competition has had many years to develop these tools, while we are new
    commers to this arena.  It is not just the box that makes or breaks the
    sale, it is the overall capability to get the most from it in the most
    cost effective, timely, labor un-intensive manner.  This I feel is the
    real problem with the 9000.  As a company, we are still operating on a
    mom and pop concept, while the competition is has a good sense of the
    scale of the customers needs.  It's kind of like the pony express vs
    federal express, both are fast, but one handles LARGE VOLUME.
    
    IMHO     Tom

    I've done my share of bitching over the years, but the rhetoric in here 
    is just plain wrong!
    
    Let's say you have a current product on the market that depends on some
    really exotic, expensive components.  And these components cost a
    *fortune* to build.  And, let's say your product has done poorly in 2
    years out on the market.  Not to mention the fact that it was about a
    year late even getting to market.
    
    Now let's say you have some other technology that looks *real*
    promising.  Turns out you can build a CPU with this new technology
    that's faster than the exotic original CPU.  And you come up with
    packaging that doesn't cost near as much as the original packaging. 
    *AND* the new packaging's I/O capabilities are still plenty good enough
    to handle the load put on them by the new CPUs and projected future
    CPUs.  *AND* the new packaging has the ability to support some of the
    new bus technology that will hit the streets over the next few years.
    
    So the new technology does just about everything the old technology
    does, and a whole lot more.  And it's less expensive to build.
    
    Would you continue with the original technology, or would you adopt the
    new technology?  What would *YOU* do?
    
    I've heard the rumors and I've also heard the facts.  Let's stop bitching 
    about the 9000; the program isn't dead, the technology is changing. 
    For the better.  And change is about the *only* constant I can think of 
    in our industry.
    
    The *facts* are good.  But the "spin control" sucks.  Please please
    please don't spread nasty rumors about this.  With these rumors,
    everyone loses - you, your customers, and Digital.
    
    - Greg Scott

>    I've heard the rumors and I've also heard the facts.  Let's stop bitching 
>    about the 9000; the program isn't dead, the technology is changing. 
>    For the better.  And change is about the *only* constant I can think of 
>    in our industry.
  

Isn't that the beauty of the VAX in the first place?  I believe that the good
ol' VAX hardware architecure has seen many many different physical 
implementations...  But a VAX is still a VAX is still a VAX no matter how 
you end up screwing it together... (OK, so you need VMS on top of it... but
you know what I mean...)

Re:<<< Note 1632.44 by HERCUL::MOSER "So what's a few BUPs between friends?" >>>

>			But a VAX is still a VAX is still a VAX no matter how 
>you end up screwing it together... (OK, so you need VMS on top of it... but
>you know what I mean...)

VMS is not the only operating system that runs on a VAX computer...

-  David

    re:  .43
    
    Greg, thanks, right on as usual.
    
    What is going on is GOOD news, yes, even perceived as such by the
    several VAX 9000 customers that I've spoken with under PID in the last 
    week.
    
    Cheers,
    Ken
    

>VMS is not the only operating system that runs on a VAX computer...
>
>-  David

*sigh*

Someday I too will learn the U word...  

    I have a site that has 4 9420's and 2 more on the way.. The systems
    have all been installed for over a year and we have had only 3
    unscheduled hardware problems... Sorta the norm for a new box, but
    for a REAL NEW box.. I personally was impressed with the 9000 from
    the get go... The training was done professionally,(the plant was
    a great idea!!) and the support I have received form CSSE was the
    best I have ever seen!! Those guys/gals are great.. We DO NOT know
    how to service them though.. As was stated in another note, we think
    this is like a MVII... This is our crown jewel so to speak and we
    should throw whatever resources or people at to make SURE it succeeds..
    If it doesn't, it was the machine's fault!!!
    
    So where's the problem?? Our MARKETING stinks BIG TIME!!!! THey
    don't what the hell they are doing and really never have.. I think
    that with the VAX line so strong from it's inception that they really
    didn't have to do much to evangelize the product... Sales from my
    perspective is just as bad...  I have not met many that I would
    put any faith in.. There are some, just far and few in between to
    justify.. 
                                                
    
    
    
    
    Joe

 Digital - DEC Restarts Stalled Mainframe Strategy
	{Computerworld 21-Oct-91, p. ?}
	{Contributed by: Butch Leitz}
   Maynard, Mass - Disappointing sales and changing customer needs have spurred
 Digital Equipment Corp. into a major overhaul of the strategy behind it's VAX
 9000 mainframe line, company officials said last week.
   On Oct. 30, DEC plans to lay out initial details on the next three
 mainframe systems slated for delivery in 1991-1993, as well as on the upgrade
 path to the Alpha VAX (sic) mainframe. Current users of the VAX 9000 will be
 offered substantial discounts and upgrade deals to move them to the new
 mainframes, said Richard Whitman, manager of DEC's mainframe line.
   Meanwhile, in hopes of boosting sale immediately, DEC last week announced
 price cuts of up to 30% on its VAX 9000 servers, and server software prices
 were cut by up to 30%.
   The company has also reorganized its mainframe business unit, now dubbed
 Production Systems Business Unit, and is shifting at least 200 employees into
 other company divisions.
   "We like DEC's game plan, but obviously, we're going to watch the execution
 of it," said George Kerns, vice president of information management at GTE
 Mobile Communications, Inc. in Atlanta. "As we triple the size of our
 processing load in the next 12 months, we want to be sure the products on the
 hardware and software side are there to help us manage that load."
   Kerns was one of several mainframe customers briefed last month on the new
 strategy. Many of the those users have been hammering DEC to give
 mainframe-class operating system software the same attention hardware gets,
 Kerns noted.
   The changing strategy is designed to do just that, Whitman said. He outlined
 for Computerworld what DEC will announce next week on the VAX 9000.
   First, DEC reorganized the mainframe business unit in a way that
 de-emphasizes hardware in favor of a production system approach, providing the
 kind of operating system software and tools that customers are demanding.
   All VAX 9000 system management software is now the responsibility of DEC
 Vice President David Stone's Software Products Group, and VAXcluster
 engineering, and development now belongs to DEC Vice President William
 Demmer's VAX Systems and Servers Group.
   DEC will also announce a VMS software products rollout during the next six
 to nine months. Featured will be data-center tools such as class schedulers,
 batch and print queue scaling capabilities and extended memory support, as
 well as other software designed to improve systems management and
 availability.
   Three new VAX 9000s will be coming between the current product line and the
 Alpha line of reduced instruction set computing machines in late 1993. They
 will be based on CMOS chips rather than the current emitter-coupled logic
 chips in what Whitman said results in "dramatic" cost savings in manufacturing
 for DEC and price/performance improvements for users (see story below).
   The first CMOS based models will be the 9x15 line, slated to be introduced
 in 1992 with enhanced memory (up to 2Gbytes internally), greater I/O and
 additional reliability features. Next summer, the VAX 9600 will arrive as the
 upgrade platform to the Alpha line. Customers will be able to board-upgrade
 the VAX 9600 to the Alpha VAX without changing software or peripherals.
   A VAX 9800 is also expected to be available in early to mid-1993 as another
 stepping stone to the Alpha VAX.
   Pricing will be disclosed in January and should stay within the current $1
 million to $3 million range, DEC officials said.
   DEC is also announcing "disaster-tolerant" clustering of VAX 9000s through
 the use of Fiber Distributed Data Interface (FDDI) links. Lloyds Bank of
 London, which has the largest DEC mainframe installation in Europe,  is 
 field-testing  a disaster-tolerant cluster of 4 VAX 9000s now.
   Alan Prout, head of systems at Lloyds Registrar's Department in Worthing,
 England, said the mainframe FDDI clustering capability allows the bank to
 divide the machines between two physically separate computer centers there.
   "We find that very attractive because we can distribute the data between the
 two centers," Pout said.
 -------------------------------------------
 Smaller, Cooler

   Imagine the difference between hefting a 300 pound engine and holding a
 matchbox car in the palm of your hand.
   That is one way to envision the dramatic physical contrast between chips
 based on emitter-coupled logic (ECL) and CMOS technology.
   For DEC VAX 9000 users, the change in the underlying chip technology, moving
 from ECL to CMOS on the mainframe, will translate into more performance at
 less cost.
   Manufacturing advances in chip packaging are enabling DEC to put on a single
 CMOS chip what took dozens of ECL chips to accomplish, noted Phil Grove,
 Marketing manager for the VAX 9000.
   While both are transistor based silicon chips, CMOS and ECL differ markedly
 in the heat and power they generate. CMOS was designed to require less power
 to complete its different logic functions, but until now it needed more space
 to perform the same function as an ECL chip. The older ECL technology, while
 more powerful, was considerably hotter than CMOS. When used in mainframes, it
 required elaborate cooling systems.

    Re .-1
    
    I'd be really curious to know whether George Kerns violated any sort
    of Non Disclosure agreement by giving all this info to the press. Or
    whether he was encouraged to do so by Digital. This is certainly the
    most detailed information I've seen to date on Digital's mainframe
    strategy. I would have expected to see this sort of info in LIVEWIRE
    before seeing it in the press.
    
    Dave

Re: .50

>    I'd be really curious to know whether George Kerns violated any sort
>    of Non Disclosure agreement by giving all this info to the press.

Whoa!  The, uh, *interesting* information is attributed to Rich Whitman at
Digital, not Kerns.  Considering the rumors that could be flying around about
the 9000, I think Rich did a good job of presenting things so that it resulted
in a very positive, upbeat article. 

    Re .-1
    
    Woops, you're dead right. I take back my question. I'm glad to see
    Digital putting out solid information to our customer base and the
    press like this.
    
    Dave