Conference 7.286::digital

    Given your versatility, you can easily read the docket filed in the
    Worcester Cty courthouse.  Bob Palmer said that the lawsuit will not be
    on the Web page.

    Other bits of news, but nothing technical, at
    http://weblib.ako.dec.com/
    
    Interesting comment from an analyst at Salomon Brothers.
    

	At least Harry Copperman used the word OpenVMS. :-)

        Question to the moderators:

        Would it be OK to allow postings of pointers to external
        information on the suit (web pages, magazine articles, etc.)
        accompanied by *no* comment?

        Bob

    Re .1:
    
    > Given your versatility, you can easily read the docket filed in the
    > Worcester Cty courthouse.
    
    I would have to be very versatile to read the docket of a United States
    District Court suit filed in the Worcester County courthouse.  But
    thanks for trying to help.
    
    
    				-- edp
    
    
Public key fingerprint:  8e ad 63 61 ba 0c 26 86  32 0a 7d 28 db e7 6f 75
To find PGP, read note 2688.4 in Humane::IBMPC_Shareware.

    Entire transcripts of the meeting with analyst has been posted in the
    Digital_Investing conference under "Digital in News" topic.  Bob Palmer
    says how one can get the detailed information about the 10 patents
    listed in the lawsuit.
    
    - Vikas
    
    P.S. You can also hear by dialing
    
    800-633-8284 International: -- 303-248-1201
    Reference # 2762376 until day after tomorrow noon

If you'll send me external documents and other public statements on the case,
I'll post them as replies to note 5284.

				Steve

    External info source
    http://www.news.com/News/Item/0,4,10612,00.html?latest

When we're done with Intel, maybe we can sue The Worcester 
Telegram for calling us DEC and not DIGITAL.

Shouldn't we prioritize that ahead of Intel? ;^)

    Shouldn't we just settle for all 'INTEL INSIDE' labels to be removed
    from all units sold to date and replaced with 'Digital Inside'??

    A similar story... kind of amusing, considering yesterday's news...
    
  Date: Wednesday, May 14, 1997 
  Source: Business Wire 
    
    RICHARDSON, Texas--(BUSINESS WIRE) via Individual Inc. -- Cyrix
    Corporation (NASDAQ:CYRX) today announced that it has filed suit
    against Intel Corporation for patent infringement in the U.S. District
    Court for the Eastern District of Texas. 
    
    Cyrix asserts in the suit that Intel products including Pentium(R),
    Pentium Pro and Pentium II microprocessors infringe on one or both of
    the patents issued to Cyrix on May 13, 1997. The suit seeks preliminary
    and permanent injunctive relief along with unspecified damages and
    fees. 

    
    
    PC Week's cut at it:
    
    http://www8.zdnet.com/pcweek/news/0512/13esuit.html
    
    Kevin
    

    Remember "loose lips sink ships"!!!

And don't count your chickens before they hatch.

Audio of press conference is available at:

http://www8.zdnet.com/pcweek/radio/0512/srzdig.ram

You'll need a RealAudio plug-in for your browser, configured with the 
appropiate proxy.  See GYRO::INTERNET_TOOLS.


						-Paul

           <<< RICKS::SYS$SYSDEVICE:[NOTES$LIBRARY]DECHIPS.NOTE;1 >>>
                                -< Hudson VLSI >-
================================================================================
Note 696.2               Intel lawsuit - which patents?                   2 of 2
ACISS2::LENNIG "Dave (N8JCX), MIG, @CYO"              8 lines  14-MAY-1997 11:09
--------------------------------------------------------------------------------
    The text of the lawsuit, including patent descriptions is at:
    
    http://www.boston.com/globe/eco/14dectext.htm
    
    re: .1 - I was not discussing the case; I simply asked for information
    that is a matter of public record.
    
    Dave

Extracted from http://www.boston.com/globe/eco/14dectext.htm

July 5, 1988, United States Patent No.  4,755,936 titled ``Apparatus And 
Method For Providing A Cache Memory Unit With A Write Operation Utilizing 
Two System Clock Cycles''

July 11, 1989, United States Patent No.  4,847,804 titled ``Apparatus And 
Method For Data Copy Consistency In A Multi-Cache Data Processing Unit''

February 25, 1992, United States Patent No.  5,091,845 titled ``System For 
Controlling The Storage Of Information In A Cache Memory,''

June 23, 1992, United States Patent No.  5,125,083 titled ``Method And 
Apparatus For Resolving A Variable Number Of Potential Memory Access 
Conflicts In A Pipelined Computer System''

September 15, 1992, United States Patent No.  5,148,536 titled ``Pipeline 
Having An Integral Cache Which Processes Cache Misses And Loads Data In 
Parallel''

January 12, 1993, United States Patent No.  5.179,673 titled ``Subroutine 
Return Prediction Mechanism Using Ring Buffer And Comparing Predicated 
Address With Actual Address To Validate Or Flush The Pipeline''

March 23, 1993, United States Patent No.  5, 197, 132 titled ``Register 
Mapping System Having A Log Containing Sequential Listing of Registers That 
Were Changed In Preceding Cycles For Precise Post-Branch Recovery''

February 28, 1995.  United States Patent No.  5.394.529 titled `Branch 
Prediction Unit For High-Performance Processors'

July 4, 1995, United Sates No.  5.430.888 titled ``Pipeline Utilizing An 
Integral Cache For Transferring Data To And From A Register ''

October 22, 1996, United states Patent No.  5.568.624 titled ``Byte-Compare 
Operation For High-Performance Processor''

    
    CNN news this morning carried the story, and used the OLD blue 
    logo.  
    
    Another suitable mistake!

Wow, sounds like FY98 could be a VERY profitable year.

    re:.-1
    
    Yeah...we may even regain #2 status  ;-)

    http://www.boston.com/cgi-bin/btally.cgi

    re: .22
    
    I just cut-and-paste the reference you supplied -- I got page not
    found.

	same here..... 

    re: missing argument
    
    http://www.boston.com/cgi-bin/btally.cgi?digital
    
    /Thien
    

    ...or just click on the "Vote" icon in the main Globe article on the
    suit.

    When faced with a similar situation didn't Microsoft buy out the
    plaintiff?

    .-1 kind of reflects what I've been thinking, that we approached Intel
    in '90 about the Alpha, and how maybe we may end up as partners (albeit
    hostile) in someway.
    
    Also, would Microsoft be next over the "similarities" in NT to VMS?  Or
    has this "battle" been fought already and we came out as strategic
    "partners"?
    
    .02

re Note 5286.28 by MAASUP::LAVELLE:

>     has this "battle" been fought already and we came out as strategic
>     "partners"?
  
        From the widely circulated reports I've read, that is in fact
        the case.

        Bob

    FWIW, if anyone is interested, IBM has a patent server at
    patent.womplex.ibm.com. A query by patent number can be done on page
    patent.womplex.ibm.com/patquery.html.
    
    The patent numbers in question are in the text of the lawsuit (I think
    a pointer was already here somewhere).
    

From:

   "http://www.techstocks.com/~wsapi/investor/s-549/reply-21822"


   4,755,936
   
   A cache memory unit is disclosed in which, in response to the
   application of a write command, the write operation is performed in
   two system clock cycles. During the first clock cycle, the data
   signal group is stored in a temporary storage unit while a
   determination is made if the address signal group associated with
   the data signal group is present in the cache memory unit. When the
   address signal group is present, the data signal group is stored in
   the cache memory unit during the next application of a write
   command to the cache memory unit. If a read command is applied to
   the cache memory unit involving the data signal group stored in the
   temporary storage unit, then this data signal group is transferred
   to the central processing unit in response to the read command.
   Instead of performing the storage into the cache memory unit as a
   result of the next write command, the storage of the data signal
   in the cache memory unit can occur during any free cycle.
   
   4,847,804
   
   In a multi-processor unit data processing system, apparatus and
   method are described for providing that only the most recent
   version of any data signal group will be available for manipulation
   by a requesting data processing unit. A "multiple" state for a data
   signal group is defined by the presence of a particular data signal
   group stored in the cache memory units of a plurality of data
   processing units. The "multiple" state is associated with each copy
   of a data signal group by control signals. When a data signal group
   is changed by the local data processing unit, an "altered" state is
   associated with the new data signal group. The simultaneous
   presence of an "altered" state and "multiple" state is forbidden
   and requires immediate response by the data processing system to
   insure consistency among the data signal groups. In addition to
   apparatus for identifying and storing the state of the data signal
   groups, apparatus must be provided for communication of the
   selected states to the data processing units.
   
   5,091,845
   
   The invention provides a system for controlling the storage of
   information in a cache memory and features a processor to be
   connected to a bus, the bus including information signal transfer
   lines for transferring information signals and a cache control
   signal transfer line for transferring a cache control signal
   having a plurality of conditions, the processor including a
   cache memory and a bus interface circuit connected to the cache
   memory and for connection to the bus, the bus interface circuit
   including: i. an information signal transfer circuit for
   performing a read operation in which it receives information
   signals from the information signal transfer lines, the
   information signal transfer circuit transferring the received
   information signals to the cache memory; and ii. a cache control
   circuit connected to the cache memory and the information signal
   transfer circuit and for connection to the cache control signal
   transfer line for controlling whether the received information
   is to be stored in the cache memory in response to the condition
   of the cache control signal.
   
   5,125,083
   
   An operand processing unit delivers a specified address and at
   least one read/write signal in response to an instruction being
   a source of destination operand, and delivers the source operand
   to an execution unit in response to completion of the
   preprocessing. The execution unit receives the source operand,
   executes it and delivers the resultant data to memory. A "write
   queue" receives the write addresses of the destination operands
   from the operand processing unit, stores the write addresses,
   and delivers the stored preselected addresses to memory in
   esponse to receiving the resultant data corresponding to the
   preselected address. The addresses of the source operand is
   compared to the write addresses stored in the write queue, and
   the operand processing unit is stalled whenever at least one of
   the write addresses in the write queue is equivalent to the
   read address. Therefore, fetching of the operand is delayed
   until the corresponding resultant data has been delivered by
   the execution unit.
   
   5,148,536
   
   A load/store pipeline in a computer processor for loading data
   to registers and storing data from the registers has a cache
   memory within the pipeline for storing data. The pipeline
   includes buffers which support multiple outstanding read request
   misses. Data from out of the pipeline is obtained independently
   of the operation of the pipeline, this data corresponding to the
   request misses. The cache memory can then be filled with the data
   that has been requested. The provision of a cache memory within
   the pipeline, and the buffers for supporting the cache memory,
   speed up loading operations for the computer processor.
   
   5,179,673
   
   A method and arrangement for producing a predicted subroutine
   return address in response to entry of a subroutine return
   instruction in a computer pipeline that has a ring pointer
   counter and a ring buffer coupled to the ring pointer counter.
   The ring pointer counter contains a ring pointer that is changed
   when either a subroutine call instruction or return instruction
   enters the computer pipeline. The ring buffer has buffer locations
   which store a value present at its input into the buffer location
   pointed to by the ring pointer when a subroutine call instruction
   enters the pipeline. The ring buffer provides a value from the
   buffer location pointed to by the ring pointer when a subroutine
   return instruction enters the computer pipeline, this provided
   value being the predicted subroutine return address.
   
   5,197,132
   
   A register map having a free list of available physical locations
   in a register file, a log containing a sequential listing of
   logical registers changed during a predetermined number of cycles,
   a back-up map associating the logical registers with corresponding
   physical homes at a back-up point in a computer pipeline operation
   and a predicted map associating the logical registers with
   corresponding physical homes at a current point in the computer
   pipeline operation. A set of valid bits is associated with the maps
   to indicate whether a particular logical register is to be taken
   from the back-up map or the predicted map indication of a
   corresponding physical home. The valid bits can be "flash cleared"
   in a single cycle to back-up the computer pipeline to the back-up
   point during a trap event.
   
   5,394,529
   
   A pipelined CPU executes instructions of variable length, and
   references memory using various data widths. Macroinstruction
   pipelining is employed (instead of microinstruction pipelining),
   with queueing between units of the CPU to allow flexibility in
   instruction execution times. A branch prediction method employs a
   branch history table which records the taken vs. not-taken history
   of branch opcodes recently used, and uses an empirical aglorithm
   to predict which way the next occurrence of this branch will go,
   based upon the history table. The branch history table stores in
   each entry a number of bits for each branch address, each bits
   indicating "taken" or "not-taken" for one occurrence of the
   branch. The table is indexed by branch address. A register stores
   the empirical aglorithm, and upon occurrence of a branch its
   history is fetched from the table and used to select a location
   in the register containing a prediction for this particular
   pattern of branch history.
   
   5,430,888
   
   A load/store pipeline in a computer processor for loading data
   to registers and storing data from the registers has a cache
   memory within the pipeline for storing data. The pipeline
   includes buffers which support multiple outstanding read request
   misses. Data from out of the pipeline is obtained independently
   of the operation of the pipeline, this data corresponding to the
   request misses. The cache memory can then be filled with the
   requested for data. The provision of a cache memory within the
   pipeline, and the buffers for supporting the cache memory, speed
   up loading operations for the computer processor.
   
   5,568,624
   
   A high-performance CPU of the RISC (reduced instruction set) type
   employs a standardized, fixed instruction size, and permits only
   simplified memory access data width and addressing modes. The
   instruction set is limited to register-to-register operations and
   register load/store operations. Byte manipulation instructions,
   included to permit use of previously-established data structures,
   include the facility for doing in-register byte extract, insert
   and masking, along with non-aligned load and store instructions.
   The provision of load/locked and store/conditional instructions
   permits the implementation of atomic byte writes. By providing a
   conditional moveinstruction, many short branches can be eliminated
   altogether. A conditional move instruction tests a register and
   moves a second register to a third if the condition is met; this
   function can be substituted for short branches and thus maintain
   the sequentiality of the instruction stream.

References

   1. http://www.uspto.gov/
Auto-saving...doneGarbage collecting...Auto-saving...done

From Usenet (so it's already well and truly public):

MAYNARD, Massachussetts, May 14 (Reuter) - Digital Equipment Corporation said 
Wednesday that it had filed a lawsuit against Cyrix Corp claiming that Cyrix had 
illegally copied Digital's plans for suing Intel in order to make up for having 
lost out to Intel in the market.

Robert Palmer, former British pop singer and now President of Digital, was  
quoted as saying, "We don't mind competing with other lawyers, but when those 
lawyers are imitating our style, well, that's just not fair."

Cyrix had no immediate comment, but analysts noted that Cyrix may agree to
a settlement with Digital, unless the legal teams hired by each company can
find some good ways to draw out the process for many years.

PJDM

Business Week, May 26th, 1997

DIGITAL VS. INTEL: IS THE PENTIUM IN PERIL? 
Not likely. After the hoopla, Digital's lawsuit may mean little 

It was a bold stroke. Not only was Digital Equipment Corp. taking on
chip giant Intel Corp. in a patent-infringement suit, it was doing 
it in a way to gain maximum publicity.  Instead of notifying Intel
directly, Digital Chairman Robert B. Palmer sprang the news at a press
conference where he intimated that the vast market power--and billions
in profits--that Intel derives from  Pentium chips is based on purloined
Digital technology. Moreover, he asserted,  with its suit Digital would
change the course of the industry. The next day, full-page newspaper
ads spelled out Palmer's position.

But Palmer says the exchange was ''very professional'' when he phoned Intel
CEO Andrew S. Grove on May 13. Grove acknowledged he had received Digital's
suit, and the two had a brief, cordial chat, says Palmer. ''Andy said, 'It
looks like one for the lawyers,''' he recalls.

In the end, this could be how the Digital-Intel battle shakes out: A lot
of public flash and a businesslike approach in private. Not that Digital
isn't going to vigorously pursue its claim--or that it doesn't have a 
substantial case to make.  It's just that such suits rarely end with the
kind of bang this started with.

For now, the experts are scrambling to understand Digital's charges.
''We haven't found any smoking guns,'' says Linley Gwennap, editor of 
Microprocessor Report, who has made a preliminary review of the claims. 
''But some things are close to what Intel's doing.'' Investors were worried 
enough to knock Intel shares down 6 3/4, to 152 3/8.

Digital's lawsuit hinges on 10 patents related to its Alpha microprocessors
that cover how data is queued up for faster processing on a chip.
Specifically, it cites how Intel chips gain speed by storing and moving
data in ways that Alphas do.

Such claims have in the past been difficult to prove.  And even if Digital
prevails in the end, Intel's juggernaut isn't about to be derailed. 
The courts are highly unlikely to grant Digital's demand that Intel
''stop using our technology.''  Rather, Palmer's best shot may be a
settlement that induces Intel to pay royalties. ''It'll be hard for
Digital to establish the validity of the claims,'' says Nathan Brookwood,
a microprocessor analyst with Dataquest Inc. ''The legal process is very
tricky in this area.''

Brookwood figures if Digital and Intel agree to royalties of 1% to 2% on
each Pentium, Digital could get some $500 million a year.  In the meantime,
however, analysts say Intel is likely to countersue Digital for infringing
its patents.

For now, Palmer brims with bravado. ''I'm certain we'll prevail, he says.''
In fact, the lawsuit may be his only way to improve the payback on Alpha,
which has consumed billions, but has never succeeded in the market 
against Intel chips.  In the meantime, Digital could invest millions in legal 
expenses only to come up empty.  And then there's the risk of strained
relations with Intel, which supplies the chips for computers that account
for one-quarter of Digital's product sales.

Palmer may want to keep his phone line open.

By Paul Judge in Boston, with Andy Reinhardt in San Francisco 

Reuters reports that Intel has sued DEC, alleging it has refused to
return confidential information.

See http://biz.yahoo.com/finance/97/05/28/dec_intc_1.html

More details at

http://www.news.com/News/Item/0,4,11005,00.html

    The battle is joined...

    (http://www.msnbc.com/news/77121.asp)
    
    

    RE: .36
    
    ... and the only winners will be the bloody lawyers...
    
    Cheers, Laurie.

    Hewlett Packard stands to be a big winner as DIGITAL stumbles
    itself out of the Intel space (90+% of DIGITAL's NT volume), which
    should pretty much eliminate the NT story DIGITAL has been trying
    to build.
    .02 K

    Actually, I think this is pretty good.  You notice it's not a
    countersuit claiming patent infringement.
    

>    Actually, I think this is pretty good.  You notice it's not a
>    countersuit claiming patent infringement.

Perhaps that suit will just take them longer than 2 weeks to gen up.    

>    Hewlett Packard stands to be a big winner as DIGITAL stumbles
>    itself out of the Intel space (90+% of DIGITAL's NT volume), which
>    should pretty much eliminate the NT story DIGITAL has been trying
>    to build.
>    .02 K

Getting out of the Intel space may not be bad after all.  Our profitability
would probably shoot up, that would please shareholders.

RE: .42

>Getting out of the Intel space may not be bad after all.  Our profitability
>would probably shoot up, that would please shareholders.

I respectfully disagree.  I believe that a Wintel (sic) desktop or notebook
is a part of 99.99999% of every sale of computer equipment today, and is
absolutely a part of every server sale, because every server sale means
that there are some clients out there somewhere, and the clients are
overwhelmingly Wintel systems.

Therefore, if our server sales force is prevented from selling Wintel
clients as part of the total solution, customers are going to look 
elsewhere for their solution.  We have a variant of that situation now,
where the SBU Sales Reps are not allowed to sell Wintel systems, but
at least they are allowed to bring in the PCBU people.  This isn't even
close to perfect, because the customer is forced to deal with multiple
people for what should be a single package, but it can work.

But if we can't sell *any* Wintel boxes, I believe this would have an
immediate and deleterious effect on our entire Sales force.  And this
would happen just when we have fixed the problem, by the merge of the
SBU and the PCBU Sales force under Bruce Claflin.

-- Ken Moreau

    
    
    Ken,
    
    although I can hardly see the fact of losing our Intel-based clients
    and servers offerings as positive, I disagree with your views on the
    possible consequences for the following reasons :
    
    - First of all, SUN seems to be doing quite well without Intel based
    systems (there is still and will be a huge market for UNIX systems, and
    Alpha-based NT is more Digital-specific stuff than Intel-based NT)
    
    - We rarely sell PCs through the PCBU, as they impose their
    distributors/resellers for the sale. In the worst of cases, we can
    strike an agreement with any Intel-based computer maker to sell their
    PCs instead of ours (when we need them), and really concentrate on Alpha
    (and keep fighting against the Intel Goliath (which I think is a winner
    mediatically, customers WANT alternatives) without any fear of ruining a
    no-longer existing relationship).
    
    However, I am not able to bet on the fact that this is the best growth
    strategy we could come up with, but, in the best hypothesis, it could 
    even strengthen our credibility as a fully Alpha-oriented company.
    
    Of course, the best would still be to keep building Alphas AND Intel
    machines, and I somehow wonder how much Intel is ready to give up a
    lucrative deal with a major customer such as us. 
    
    Just my opinion.
    
    Michel.  
                       

    In the answer Digital did or will file with the court:
    
    	Does Digital accept or deny that Intel had a right to
    	require the return of materials?
    
    	Does Digital accept or deny that Digital returned the materials?
    
    
    				-- edp
    
    
Public key fingerprint:  8e ad 63 61 ba 0c 26 86  32 0a 7d 28 db e7 6f 75
To find PGP, read note 2688.4 in Humane::IBMPC_Shareware.

    re: .45
    
    >Did Digital return them?
    
    According to published reports, the answer to this is "no".
    
    Of course, whether or not such published reports are accurate is an
    unresolved question.
    
    The local paper (Baltimore Sun) carries a report this morning that
    Intel claims they have no long term agreement with us w.r.t. chip
    supply.  This flies in the face of the official response to the Intel
    suit which claims that we have some long term agreement with Intel for
    chip supply.
    
    I've noticed, though, many articles reference us as having "the fastest
    chip in the world" in Alpha.  This may be the most effective means of
    advertising we've ever had.  8^(
    
    -- Russ

    Regarding INTEL chip supply from a BLOOMBERG buisness report:
    
    
    Intel has no contract to supply DIGITAL with processors after the end
    of the third quarter."Clearly we will honor all contractual
    agreements,"although there are none after the end of September,said
    Intel spokesman Howard High.
    
    Intel has a long-term supply agreement to sell Digital other kinds of
    chips, such as embedded controllers and flash memory chips, through
    June 30,1999.
    
                                 Rich
    
      

    Is Pentium second sourced?  Can you get Pentiums from other sources
    than Intel?  The article in today's Globe seem to say that.
    
    				-John

    Not "second sourced", but there's wholesalers resell them at a markup.

|    Not "second sourced", but there's wholesalers resell them at a markup.

I cannot imagine that basing your product line on "spot market supply" is
very viable in volume. Getting other motherboards from someone like Gateway
or Dell may be more dependable.  

>Per

    
    	RE: .48
    
    	There are "Pentium alternatives" offered by other CPU manufacturers,
    	if that counts.  Cyrix, possibly AMD.
    

>I cannot imagine that basing your product line on "spot market supply" is
>very viable in volume. Getting other motherboards from someone like Gateway
>or Dell may be more dependable.  
    
    I think both Dell and Gatway use Intel motherboards.

    

    re: Note 5286.51 by BUSY::SLAB
    
    We've recently announced that we're going to use the AMD K6 in some
    PC lines. Perhaps this move was in anticipation of an Intel retaliatory
    move. 
    
    Ken....

    
    Re AMD K6.
    
    But, Bob has stated publicly that he didn't anticipate ANY retaliatory
    action by Intel.
    
   ed

    I've heard the problem with the AMD K6 is in SMP.  True of false?
    
    You can always buy someone else's product and slap our logo on it like
    we did with HP laser printers.

The AMD K6 does not support SMP.  As an AMD spokesman explained it to me (I saw
a presentation at a Microsoft DevDays show), Intel has patented the method
Pentiums use for doing SMP processor selection and AMD doesn't have access to
it.

				Steve

    True, K6 doesn't do SMP.
    
    (DIGITAL's, not just the PCBU's, largest customer, the US Post Office
    Prioris server contract, uses SMP Intel-based Prioris servers, for
    example.)
    K

    Maybe we can help them achieve it ;-)

RE: .56

>The AMD K6 does not support SMP.  As an AMD spokesman explained it to me (I saw
>a presentation at a Microsoft DevDays show), Intel has patented the method
>Pentiums use for doing SMP processor selection and AMD doesn't have access to
>it.

   That's not what has been discussed on the newsgroup comp.arch.  According
to what I've seen there, the K6 will do SMP, but not Intels way. From what
I've read there, Intel uses a protocol called APIC.  Since they don't document
this, Cyrix and AMD had to come up with their own protocol called OpenPIC.
This would require a different motherboard design in order to utilize the
SMP capability.  But the chip doesn't preclude SMP at all.

Bill Powers

RE: .44

Michel, thanks for the substantive reply.  You make some very good points,
which I happen to disagree with.

First, you state that Sun seems to be doing well without an Intel line.
Good point, but in the deals where I have been competing with Sun, they
always walk in arm-in-arm with Compaq to fill that void.

Second, you state that we rarely sell PCs direct, but depend on resellers
and distributors for order fulfillment.  Another good point, but this is
where I think your argument falls down.

Sun can work with a PC distributor/reseller/direct_vendor, because they
don't have an Intel line, and therefore their partner does not compete
with them.  Sun does Solaris, the partner does Wintel, and everyone is 
happy.  But in our case we have the Alpha doing NT, which competes with
the partner.  So if we brought a partner into an account with us, they
would (naturally, the same as we do) want to maximize their revenue in
that account, and so would in fact compete with us.  So the customer would
have one of his vendors selling Wintel servers and high-end desktops,
and the other vendor selling Alpha servers and high-end desktops, sometimes
with the same Microsoft-supplied OS.  This would cause some confusion :-(

Finally, I think Digital management would have a great deal of difficulty
inviting a Compaq or a Dell into a Digital account, because this little
mis-understanding with Intel won't last forever, and we will at some time
in the future offer Intel systems again, and it would be exceptionally
difficult to unseat a Compaq or a Dell, after we invited them in...

Your final point (I wonder how much Intel is willing to give up with a
major vendor such as Digital) I believe is slightly myopic and inward
focused.  I believe that Intel believes (and FWIW I agree with this) 
that the market for business PCs is in-elastic, the Wintel PC is a
commodity, and that customers who want PCs will simply shop around 
for a vendor to buy them from.  From Intel's point of view, if Digital
sold 'n' million less PCs next year, customers would simply buy exactly
'n' million more PCs from Compaq or Dell or HP or IBM or ..., and Intel's
revenues would not be disturbed in the slightest.  So from Intel's
point of view, who cares if Digital sells *any* PCs next year?

-- Ken Moreau
   who is glad for the abundance of light and lack of heat in the last
   series of replies

    Re: .59
    
    All I know is what the AMD guy said when asked a direct question on
    SMP support - this was back in March.  Perhaps the story has changed
    since then (or he didn't know or didn't want to say).
    
    					Steve

    
    Fwiw, my $.02 is that Digital and Intel are to tightly joined at the
    hip for a long drawn out battle. For those who doubt that Intel needs
    us as bad as we need them, consider this :
    
    A recent consulting engagement at Intel came across a bunch of red neck
    manufacturing users who refused to go migrate to the proposed new mail 
    system. Guess what mail system these folks are on ?
    
    VMSmail. Their manufacturing operation is run primarily on OpenVMS.
    
    Now, is this ironic ?
    
    :-)
    
    / Kerry

From the WWWeb:

(http://www8.zdnet.com/pcweek/news/0526/28elett.html)

Following are excerpts from letters between F. Thomas Dunlap, Intel vice 
president and general counsel, and Thomas C. Siekman, Digital vice 
president and general counsel.

May 15 letter from Dunlap to Siekman: In the past few days several of your 
senior executives have been quoted as saying that DEC has "long-standing 
supply agreements with Intel."  We are not aware of any supply agreement 
between Intel and DEC with respect to Pentium, Pentium Pro or Pentium II 
processors other than our usual purchase order and acknowledgment process, 
which only covers sales through Q3.

May 20 letter from Siekman to Dunlap: We are surprised by your inquiry 
regarding ...  long-standing supply agreements between Digital and Intel 
...  The current BOA [Basic Order Agreement] was extended last year and 
will now expire on June 30, 1999.

Digital senior executives have also expressed their intent that the patent 
infringement suit ...  remain a technology dispute and that it not impact 
the commercial relationship between our two companies.

...  [Since] the filing of the lawsuit, Intel representatives have canceled 
several meetings with Digital personnel, including long-arranged meetings 
between engineers to review data connected with the "Deschutes" product 
introduction.  Intel representatives ...  have requested return of 
"Deschutes Yellow Book" materials.  Ongoing review of this Yellow Book data 
is necessary to facilitate delivery of a Digital Deschutes-compatible 
product at the time of the official Intel product introduction.

We assume that these recent actions are aberrations ...

May 27 letter from Dunlap to Siekman: Intel is amazed that DEC continues to 
state that it expects the relationship to remain the same as it was before 
DEC filed the 10 patent lawsuits ...  DEC's top executives went so far as 
to publicly allege that Intel has "copied" and "stolen" DEC's technology 
...  It is unreasonable to think that the relationship between the 
companies can remain the same in light of these unfounded public statements 
attacking Intel's integrity ...

You apparently believe that Intel will continue to transfer its technology 
to DEC by continuing to provide Intel-confidential technical information 
...  These Intel assets are valuable Intel intellectual property and we 
certainly do not intend to provide them to DEC while DEC is suing Intel on 
DEC's intellectual property and accusing Intel of stealing its technology.

Accordingly, your assumption that Intel's request that DEC return Intel's 
technology is an "aberration" is completely wrong.

    as Andy Grove was quoted as saying ...
    
       "looks like it's one for the lawyers"
    
    personally, I find the first round a bit of a yawn.  I know very little
    about law but the tone of these letters doesn't inspire me.  
    
    How do these company letters get onto the public web?

    I believe they were filed with the Intel lawsuit.

    The letters sound like Alexander J. Botts and the Earthworm Tractor
    Company form the old Saturday Evening Post. Lots of huff and puff, and
    neither one listening to the other. 
    
    I hope we come out as well as Botts always did.

>               <<< Note 5286.55 by PHXSS1::HEISER "Maranatha!" >>>
>
>    I've heard the problem with the AMD K6 is in SMP.  True of false?
>    
>    You can always buy someone else's product and slap our logo on it like
>    we did with HP laser printers.

For the record, Digital has NEVER rebadged an HP laser printer as a Digital
product.  Similar looking laser printers in the HP and Digital lines
were based on common laser printer engines from Canon, augmented with
controllers and communications interfaces provided by HP and Digital
(or their suppliers), respectively.  HP and Digital laser printers
have always been fundamentally different machines.

Digital DID rebadge/remarket HP pen plotters and ink jet printers 
in the mid-80s, but NEVER lasers.

- tom] (in Digital laser printer development since 1983)

	What's this "DEC" company? I thought Digital..er..DIGITAL..er..
	sued Intel?

						mike

|Digital DID rebadge/remarket HP pen plotters and ink jet printers 
|in the mid-80s, but NEVER lasers.
    
    Thanks for the info, I probably confused the two.  I think the LVP16
    was one of the rebadged HP plotters.

    
    I can't believe it that Palmer made mistake that we have
    long term contract with Intel? 

    From the Wall Street Journal:   International Business Machines Corp.
     (IBM) Friday introduced the IBM 6x86MX, a Pentium-compatible
     microprocessor that gives customers the same performance as Intel
     chips, but at a lower price.
    
    Me on Intel huffing and puffing: yawn.  Does anyone really think
    that Dec filed this lawsuit without knowing exactly where we stand
    in terms of chip availability, etc. from either Intel or other
    suppliers?  Let's not confuse posturing for public relations with
    substantive issues and beliefs.
    

    
    	No, DEC never makes an error.
    
    ed

RE: .72

	We make our share of errors, but we DO have a very good group
	of lawyers and our nature as a company weighs on the 
	cautious side of litigation. We wouldn't get into this without
	knowing what we are doing. I wish I could say the same for
	marketing.

							mike

    
    	RE: .72
    
    	DEC never MADE an error.
    
    	Digital Equipment Corporation, or DIGITAL, might very well make an
    	error or two.
    

Re: .71

That's actually a Cyrix chip, manufactured by IBM.

			Steve

    
    It was reported Cyrix chip is half the price of Pentinum II.
    Also today Intel stock is down on their lower estimated earning
    prediction on second quarter.

    re: Note 5286.70 by LABC::RU
    
    > I can't believe it that Palmer made mistake that we have
    > long term contract with Intel?
    
    What makes you think BP made a 'mistake'? 
    
    Ken....

    re:.43
    In fact, about 2% of PC sales go into existing Customer Base. Thus, it
    would seem that it is not important to have a PC business to help out
    with other sales.
    It was the intention of DIGITAL when it got into the PC business that
    PC sales would augument other sales and thus help to secure sales that
    might have been in the balance. This has long been forgotten as we have
    tried to compete with the likes of Dell, Compaq, IBM, etc... in the PC
    business space when all along that was not the primary business intent
    of PC's.
    Now, DIGITAL does not really count for much more than 1.5% or so of PC
    sales world-wide, so the impact to Intel in terms of DIGITAL purchases
    is not very large. Probably spot purchases are OK for our supply
    requirements. If not, if Intel were to try to dry up the broker
    business too, then that effort, coupled with ceasing to sell product to
    DIGITAL would most likely play into DIGITALs case would smack very
    heavily of unfair trade practises and thus in any judiciary eye, they
    would come under suspicion easily on other fronts and well, maybe I am
    thinking too loud here, ...
    
    rtt

>    In fact, about 2% of PC sales go into existing Customer Base. Thus, it
>    would seem that it is not important to have a PC business to help out
>    with other sales.

I will venture that you're refering to direct sales to our customer base. 
Since most of our Intel products flow through resellers the % may be much 
higher.

The way I see thing now (and things can change), with Alpha not being able to 
gain any presence, not having access to the latest Intel technology will 
significantly constrain any significant future product revenue growth (over 
several years), IMHO.

But the market could adopt openly Intel clones, the same way it did with PC 
clones, or someway/somewhow Alpha will finally make it. As we say in spanish, 
dreams don't cost anything.

    Here's what one customer on CompuServe is wondering...
    
     Hi, I read about recent war between Intel - DEC. Do DEC plan to release 
     new PC using future Intel CPU? Do Intel support upgrade for Intel based 
     DEC PC?
    
    
    
    					Steve

The PC market and the microprocessor markets are substantially different.
Anyone with a workbench and a screwdriver can put together PC's from
fairly standard parts to make clones.  Assembling microprocessors takes
a billion dollar fab line & either license to a design or a whole ton
of engineering work to clone a design.  AMD and Cyrix (and another?)
are the only two trying to clone x86's.

>The PC market and the microprocessor markets are substantially different.
>Anyone with a workbench and a screwdriver can put together PC's from
>fairly standard parts to make clones.  Assembling microprocessors takes
>a billion dollar fab line & either license to a design or a whole ton
>of engineering work to clone a design.  AMD and Cyrix (and another?)
>are the only two trying to clone x86's.

Reason why I made the last statement in my .79 reply

                       CBQ ANALYSIS: Intel Corporation
            Digital Equipment and Cyrix Patent Infringement Suits

EE Times Article

http://techweb.cmp.com/eet/news/97/957news/chief.html



                         Digital chief girds for battle

                                By W. David Gardner

MAYNARD, Mass. -- A month ago, Digital Equipment Corp.'s Robert B. Palmer was
among the most visible men in the computer industry, propelled into the
spotlight by Digital's stunning patent suit against Intel Corp. Today, the
bunkers are going up at Digital headquarters here as forces on Wall Street and
Main Street--and, of course, at Intel--circle the embattled chairman and chief
executive. 

"Mr. Palmer has been associated with the current state of affairs at Digital,"
said Wall Street shareholder activist Herbert Denton, who noted that the current
state of affairs appears dismal. Digital has lost more than $5 billion in recent
years, and its stock plunged from $199 a share to the mid-30s recently. 

Denton, president of Providence Capital Inc. (New York), will preside over a
meeting representing more than 35 percent of Digital shareholders on June 18.
Denton said the meeting is for "informational purposes." But many of his past
informational meetings have led to fundamental structural changes in companies
whose stocks have performed poorly. 

The meeting, closed to the press and the public, will cover several topics but
will zero in on Digital's lawsuit against Intel. Two intellectual-property
experts will discuss the suit and Intel's countersuit against Digital. 

And the litigation seems destined to escalate. Digital has indicated it could
file antitrust charges against Intel, and intellectual-property experts expect
Intel to take additional legal action against Digital. Said Denton: "We're
trying to understand this." 

Palmer has said through a spokesman that he will not attend Denton's meeting. 

But another gathering--Digital's annual meeting, in November--could command his
attention. The Communications Workers of America labor union and the Investor
Rights Association of America have filed separate shareholder proposals that, if
approved, could lead to Palmer's ouster. 

"We feel Digital should be merged or sold," said William Steiner, founder of the
Investor Rights Association of America (Great Neck, N.Y.). The group has filed
three resolutions for consideration at the November annual meeting: first, to
hire an investment banker to merge or sell the company; second, to remove
so-called poison-pill structures at Digital that make it difficult to remove top
management; and third, to facilitate the election of a new board of directors. 

Noting that Palmer took over at Digital in 1992, Steiner said, "Five years is
enough time for a company to get on its way." 

The Communications Workers of America (CWA) , which owns some 48,000 shares of
Digital stock, asks in a resolution for the annual meeting that Palmer's
position as chairman and chief executive be divided into two posts. A CWA
spokesman noted that more than half of Digital's employees--more than 60,000
people--have been fired in recent years. He said that more than 35 percent of
Digital's European employees are union members and that the CWA had been
consulting with its "counterpart unions in Europe" over the Digital situation. A
German Digital employee filed a proposal requiring Digital to include an
employee on its board. 

Palmer's future rides in large part on the sales performance of Digital's
flagship 64-bit Alpha processor. Denton said the June 18 meeting will feature a
presentation on the Alpha by an independent processor architect who designed one
of IBM Corp.'s most successful machines while he was employed at IBM. 

While sales of the Alpha dropped 2 percent in Digital's last reported quarter,
Alpha's fans consider the dip an aberration. In fact, sales volume was up 2
percent after a sharp price reduction. 

"I remain confident that the Alpha will be successful," said Terry Shannon of
Ashland, Mass., publisher of the newsletter Shannon Knows DEC. An upcoming, more
robust version of Windows NT could be crucial to that success. 

Shannon said Windows NT release 5.0--the so-called Cairo release, due out next
year--is a "bet your company" event for Palmer. Indeed, ever since Bill Gates
pledged to support the Alpha in 1992, Digital has been gambling that NT-powered
Alphas would provide the major forward thrust for the company. 

But the early NT versions, released in 1993, where "weak and crude," Shannon
noted. Microsoft didn't suffer much; Digital did. 

The good news for Digital, Shannon said, is that "the Alpha will be the only
64-bit computer that NT will be supported on [initially]." The problem is that
Digital's NT lead on the 64-bit Alpha could shorten significantly when Intel and
Hewlett-Packard Co. bring out their 64-bit Merced processor. The later Windows
NT 5.0 is delivered to market, the more tenuous Digital's 64-bit lead becomes. 

Digital's NT-centered relationship with Microsoft has the overtones of a
detective thriller. Digital was hardly pleased when Microsoft lured away David
Cutler, a key developer of Digital's pathfinding VMS large-systems operating
system, who went on to mastermind NT. Hawks among Digital's top management ranks
talked lawsuits; Palmer instead hammered out a peaceful pact with Microsoft. He
likewise crafted an arrangement with Oracle Corp. that saw Digital largely
vacate is database products and turn over database development for Digital's
equipment to Oracle. 



Palmer, an engineer from Digital's semiconductor unit, has strong emotional ties
to the Alpha, having initially proposed the device to Digital's directors board
in 1988. But in spite of the Alpha's eye-popping hardware specifications, its
applications software--or lack thereof--has been criticized for years. 

Digital believes it has solved the applications problem with its FX!32
binary-code translator. "This is excellent stuff," said Shannon. "Even if you
can only get 60 percent of native performance [with the FX!32 on an Alpha],
you're still ahead of Intel's top machines, because the Alpha is twice as fast." 

So why hasn't the hardware sold well? Shannon cited the perennial user complaint
about Digital: poor marketing. That translates into a poor "ease of doing
business" with the company. 

"Digital's marketing . . ." Denton mused, trailing off. "The marketing is a
fuzzy issue. We have no real clue as to
what's going on." He hopes some clues will emerge at his June 18 informational
meeting. 

Confusion about market direction has reigned at the company for years. Long
before Palmer took over, realignments and reorganizations of Digital's sales
operation were virtually annual affairs. Palmer, too, has repeatedly realigned
Digital's sales efforts, frequently forcing out its marketing executives. The
list of the disenfranchised includes sales and marketing chief Edward Lucente
(out in 1994); Bernhard Auer, head of Digital's PC unit (1995); and Enrico
Pesatori, vice president of the computer systems division (1996). 

Indeed, some of Palmer's enemies have quipped that it will be difficult to
replace him because he's never groomed an heir apparent. 

Earlier this spring, however, Palmer elevated IBM veteran Bruce Claflin into the
de facto No. 2 position at Digital. With that move, Palmer seems to "have his
ducks in a row," Shannon said, noting that Claflin whipped Digital's PC unit
into profitability and is moving to shape up its sales and marketing
organization. 

Palmer acknowledges having made mistakes. And his supporters maintain that after
five years of desperate top-level executive reorganizations and futile forays
into the treacherous PC market, Palmer finally appears to have stabilized
Digital's revolving-door management and to have brought a measure of
profitability to the company's longtime Achilles' heel: its PC business.
Meanwhile, its semiconductor operation is growing, with more than 300 customers,
and its Internet unit, paced by its AltaVista search engine, is likewise on
track. 

A Digital spokesman said the company has progressed from the point five years
ago when it was "hemorrhaging money." He said it communicates with its
shareholders and thus sees no need to attend the Denton meeting. 

So, what does Denton hope to achieve? Providence Capital is "a broker/dealer,"
he said. "We have an institutional client base. We like to make money, and we
like cheap stocks. Institutional investors have decided the company is bad." If
Digital can improve its business, then its stock will rise. Denton said the
choice is between the "status quo" at Digital and its willingness and ability to
make "some kind of deep-rooted cultural change." 

And tiny Providence Capital, which has just four full-time employees, has been
successful in pressuring some large firms--including Comsat Corp. and California
Microwave Inc.--to make structural changes. 

Texan Palmer, with his signature double-breasted suits and Porsche, is the
opposite of Ken Olsen, the CEO he replaced--a native New Englander who drove a
Ford Pinto because it saved on gas. 

Palmer now is close to where Olsen was five years ago, with restless investors
and anxious employees. But publicly, he seems confident that he won't have to
trade down that Porsche.