NetNews Usenet Archive 1992 #16/NN_1992_16.iso/spool/sci/physics/11725

Text File | 1992-07-27 | 15.7 KB | 363 lines

Newsgroups: sci.physics
Path: sparky!uunet!usc!snorkelwacker.mit.edu!thunder.mcrcim.mcgill.edu!sifon!news
From: [e-mail]
Subject: SSC cost
Message-ID: <[e-mail]>
Sender: [e-mail]
Nntp-Posting-Host: mafalda.math.mcgill.ca
Organization: McGill University
Date: Mon, 27 Jul 1992 16:24:56 GMT
Lines: 351

I pulled this off talk.politics.misc. I am not responsible for it's contents.

Path:
sifon!thunder.mcrcim.mcgill.edu!snorkelwacker.mit.edu!usc!zaphod.mps.ohio-state
edu!cis.ohio-state.edu!ucbvax!UTARLG.UTA.EDU!B645ZAW
From: [e-mail] (Stephen Tice)
Newsgroups: talk.politics.misc
Subject: Re: Supercollider "For" or "Against"
Message-ID: <[e-mail]>
Date: 27 Jul 92 08:04:00 GMT
Sender: [e-mail]
Lines: 338


 /                                                       \
(   We live in a Newtonian world of Einsteinian physics   )
<   ruled by Frankensteinian logic.                       >
(                                                         )
 \                                  -- David Russell     /


{Re: SuperCollider "For" or "Against," talk.politics.misc}

In article <1992Jul26> [e-mail]
|(Phillip M. Hallam-Baker) writes:
|
|   Perhaps Mr Tice might admit defeat in his attempts to inflate the
|   price of the SSC so that we might start a sensible discussion on
|   the basis for which such a seeminlg ludicrous sum as $5 to $10
|   billion should be spent?

Not hardly. I'd rather watch you backpeddle some more =Doctor=
Hallam-Baker; your speculations have been most entertaining. It
should be quite interesting to see what you do with real facts, as
a "researcher" in the field, that is.

I wrote on 23 July...
}
}   What they aren't saying in the Dallas Morning News is that it
}   cost about $8 billion to construct the SSC over 8 years, that
}   it has about a 25 year projected lifespan and that it will
}   take about a billion dollars a year to run.

PHB responded, in article <[e-mail]>...
|
|   Thats because it's not true.


We shall see, here are the facts:

{ == ITEM 1 == "The cost to construct by 1999, is about $8 billion." }

Supported by Congressional Quarterly Weekly Report, (6-20-92) p.1784...

    Cost: Early estimates of $4.4 billion to build the collider
    have now climbed to $8.3 billion. And critics claim the final
    price is likely to jump significantly higher. These construction
    estimates do not include the full cost of the sophisticated
    detectors and operation expenses that would be incurred once
    the collider is built....

Also, from HEARING before the SUBCOMMITTEE ON ENERGY AND DEVELOPMENT,
of the COMMITTEE ON ENERGY AND NATURAL RESOURCES, UNITED STATES SENATE
(4-16-91) "DEPARTMENT OF ENERGY'S SUPERCONDUCTING SUPER COLLIDER PROJECT"
S. HRG. 102-65, p.24...

    Senator Ford [of Kentucky]. Let me try some questions and maybe
    I can move into the arena with you, [Mr. Moore]. The total cost
    of the SSC done by your office of independent cost estimate, it
    is called ICE, is $11.8 billion. The ICE report states that "DOE's
    estimate for the SSC is both unrealistic and unachievable," and
    that is a quote.

    The report also states that it does not reflect a worst case
    scenario, that is $11.8 billion does not. An estimate for a worst
    case scenario could be substantially higher than $11.8 billion.

    Now we keep accelerating. We started out at less than $4 billion
    and then we went almost to $8 billion, and now we are at $11.8
    billion, and we just keep moving, and it keeps eating into the
    amount of monies available for basic research or whatever around
    the country.

    So the difference between your estimate and that of your own
    independent staff is $4 billion. Now explain that large difference
    to me.

    Mr. Moore [Deputy Secretary, Department of Energy]. Mr. Chairman,
    I will try.

    Senator Ford. That comes out of your group, you know, ICE does. That
    is not ours.

    Mr. Moore. This book explains it in detail, to the very word. There are
    not secrets in this, no mysticism, no lack of understanding ---

    Senator Ford. Well, how accuarate is ICE's prediction rates?

    Mr. Moore. We think -- we had three studies done, and we do not think
    that they are any more accurate than the others. We think they have
    made some mistakes.

    For instance, all the groups, the major difference between what the
    laboratory said this was going to cost and the others wass in contin-
    gency. How much money do you put aside in case something does not
    come in like you want it?

    And what we are telling you right now, so far we are coming in with
    a contigency we have selected at that or below, have not had to use
    it as a matter of fact, any of it. So we think our estimates are
    better.

    Number two, they have added some things to it, substantial amounts
    of money that were really not part of the total project costs. They
    simply say, we disagree with your policy, we think it should.

    Give you an example, they include in here the DOE site office. We
    have never included that in any project anywhere in the country,
    and when we put Government people in the ground to manage it, that
    is part of the cost. That is part of the overhead of running Depart-
    ment of Energy. It has never been allocated to a project.

    So we disagree with them on some things like that, they are just
    out of the ballpark ---

    Senator Ford. That could not be a whole lot, could it?

    Mr. Moore. Well, it means, that particular figure was not but $31
    million, but they say the cost of land ought to be in there. Texas
    is giving us the land, all of it, no matter what it costs. So what
    difference does that make?

    Senator Ford. You ought to take it to the Taxpayers' League and let
    them figure out what you are doing. They want to factor everything
    in, and so that is part of the cost to the taxpayer, whether it is
    Texas or up here.

    So you are excluding $1 billion from Texas, is that right?

    Mr. Moore. No, we are excluding the cost of land in Texas.

    Senator Ford. Well how much is that?

    Mr. Moore. Well it depends on what it winds up coming in at. Texas
    agreed to furnish it, no matter what it costs.

    Senator Ford. So the cost, ICE has reflected in ---

    Mr. Moore. They say $70 million.

    Senator Ford. Yes.

    Mr. Moore. Is what they say. We think they are off, it is going to
    cost more than that.

    Senator Ford. Well, say it is $100 million, but you do not count that
    in your estimate?

    Mr. Moore. There is no point in counting it ---

    Senator Ford. Are you going to count foreign money?

    Mr. Moore. It depends. The foreign money that is given towards the
    building of the project, yes. That is part of the $8.249 billion
    cost. In some cases, detector costs are not counted and never have
    been counted in projects because nobody knows what these things are
    going to cost, and many of them are not even finished until the project
    is finished.

    Senator Ford. But the taxpayer will pay for that out of the budg-
    etary process, to the Department of Energy?

    Mr. Moore. They will only pay what we have asked for and no more than
    that.

    Senator Ford. Oh, I understand, $533 million?

    Mr. Moore. No. It is more than that. It is -- let us see, detectors
    are what? $910 million.

    Senator Ford. How many agreements have you been able to work out and
    how much foreign contribution so far? Will it be in cash or in kind?

    Mr. Moore. The only agreement than has been worked out so far is with
    Korea. The value they intend -- excuse  me, with India, and the amount
    that they intend to contribute is in the range of $50 million.

... And so on for many more pages.

As a taxpayer, a Texan, and an American -- It's pretty clear to me I'm
getting screwed. Other government costs which arise because of the
SSC aren't included; Texas pays part (my taxes); the detectors aren't
figured in to the $8 billion; foreign nations pay -- maybe. So the
tally on cost, assuming the lower DOE estimate, comes to $8.3 billion +
$910 million {detectors} - $50 million {India} + operating cost with no
contingency padding = ??  On to operating costs.


{ == ITEM 2 == "It will take about a billion dollars a year to run." }

From an earlier Senate Hearing, S.HRG. 101-673, (4-24-91), p. 362,
Questions from Senator Ford, Answered by DOE...

    Question 3: What is the estimated operation cost of the facilty?

    Answer:     The estimated annual funding requirements for normal
                operation after completion of construction is $313.3
                million in FY 1991 dollars. The projected costs for
                laboratory operations, capital equipment, general
                plant projects and accelerator improvements are
                included.

I'd sure like to have the breakdown of DOE's projection. Instead, I may
try to get hold of a copy of UTA's Business School projection.

So the costs are $8.3 billion + $910 million (detectors) + 25 years*$313.3
million is about $17 billion, with no contigency padding, or inflation
adjustments.  YES -- 25 years.

{ == ITEM 3 == "It has about a 25 year projected lifespan." }

Now since there is some doubt about that 25 year lifespan figure, another
cite from the Senate Hearing on 4-16-91, p. 88, again DOE responding to
another set of Chairman Ford's questions...

    Question 12: Based on experience with other particle accelerators,
                 what is the expected useful life of the SSC?

    Answer:      The SSC is expected to have a useful life of at least
                 25 years after commencement of operations. All the
                 major systems and facilities, conventional construction,
                 and support infrastructure facilities are being designed
                 with this time period in mind. The lifetime projection
                 for the SSC is consistent with the operating history of
                 the three major DOE High Energy Physics (HEP) facilities.
                 The Brookhaven Alternating Gradient Synchrotron (AGS)
                 facility has been in operation over 30 years, the
                 Fermilab main ring for nearly 20 years, and the
                 Stanford Linear Accelerator Center (SLAC) linac for 24
                 years; all are still productive facilities. SLAC, Fermi-
                 lab, and Brookhaven are three DOE-supported major accele-
                 rator centers that were put in operation in the 1960's
                 and 1970's.

Now if the other governmental costs cited in the ICE report  =or=
contigencies occur =or= dollars are not adjusted to '91 equivalents...
the cost of the SSC will grow from ** $17 billion ** up, over it's
projected lifespan. Since I don't have the business school's estimate
in front of me, I don't know what the assumptions they made were. But
I do know that DOE's figures are twiced what the public is being told,
by the media, the executive, and many in congress -- according to the
record.

SSC funding is coming up for floor discussion in the Senate again,
likely this week. The Senate Appropriations Committee has already
approved, so it goes before the whole Senate now. It should be
interesting to see who's really for the taxpayer.

As for PHB, you be the judge of what to make of Doctor Hallam-Baker's
guesses, and research skills. {Rest assured PHB, there's plenty more
where this came from. Did you really think that I would rely on only
one source -- especially The Nation -- especially when I noted that I
selected it just to spin you up.} Let's see if I can't add a little
torque to the spin.

== Yet Another Item == "Terabytes in nanoseconds..."

I may type in a fuller treatment of the data-handling problem later in
the week. {Mr. Austern, will also want to take note.} I said that the
figure being "bandied about" is a "Terabyte per nanosecond," and
submit the following teasers. From the Conference Record of the 1991
IEEE Nuclear Science Symposium and Medical Imaging Conference, Nov.
2-9, 1991, Santa Fe, New Mexico, USA, vol 2 of 3...

  "Data Acquisition Studies fro the Superconducting Super Collider"
  by Milner, Booth, Botlo, Dorenbosch, and Wang of the Physic Research
  Division of the SSC, and Idate, Kappor, and Raj of the University of
  Texas, p. 929:

    The collider will provide a beam crossing every 16 ns, with a mean
    of 1.6 collisions per crossing (100 mb cross section) occuring at
    the design luminosity of 10^33 cm^-2 s^-1, giving a 100 MHz inter-
    action rate. We can expect data for each event to be about 0.4
    Mbyte, giving a raw data rate of 40 Tbyte/s...


Perhaps Doctor Hallam-Baker would care to translate that, or he might
wish to refer to another paper from the conference...


  "High Speed Data Transmission for the SSC Solenoidal Detector," by
  Branko Leskovar of Lawrence Berkeley Lab, p. 828:

    Abstract. High speed data transmission using fiber optics for the
    Superconducting Super Collider solenoidal detector has been studied.
    The solendoidal detector system will consist of nine subsystems
    involving more than 10^7 channels of readout electronics. Conse-
    quently, a new high performance data acquisition system, incorpora-
    ting high-speed optical fiber networks, will be required to process
    this large quantity of data.

    ... Using an open architecture, the data acquisition system will be
    capable of processing data rates of several thousands Gbit/s from
    detector elements to the online processing system....


And rather more amusing, though no less thought provoking, was the article
out of COMPUTERWORLD from (7-31-89), "Collider's colossal computing
conundrum," p. 4:

    ...The lab expects 100 million collisions per second.

    ..."Each time particles collide, something happens. But it's not
    necessarily something you're interested in," Gilchriese said.
    [Gilchriese was then associate director of the physics research
    division of the supercollider lab.] He said he expects reductions
    to be made from 100 terabytes "to 10, or 100 [megabytes] at most."

    No on yet knows how that filtering will be done. Artificial
    intelligence programming will not fit the needs of filtering
    information, although it can be used in machine control. Gilchriese
    said the lab's researchers were exploring neural networks to make
    heuristic decisions....


I'm tempted to go pull back up your speculations =Doctor= Hallam-Baker,
to compare with the material I've presented thus far. ;^)

I think you'll agree though, that the matter merits a =considered=
rebuttal now, regarding professional reputations that is. With all
those citations I provided previously on Perot, did you really think
I hadn't done my homework on this.

I particularly got a kick out of your trying to focus the "small
science" folk on believing that the SSC would take money from them
-- when the administration has asserted funding will all come from
new money. Shall I cite the American Physical Society's concerns?

No wonder the US is $4 trillion dollars in debt.

By the by, there's an old rule of thumb I learned doing defense
engineering work -- life cycle cost is 3 times R&D. My ballpark guess
is that the collider should cost something like $30 billion total over
it's lifespan. Of course, one doesn't care about constant dollars in
that case. And lastly, science being what it is -- there's nothing
which guarantees that the physics investigations the SSC is intended
for could actually yield any wothwhile results.

 /                 \     |     /  Of all sciences, there is none where  \
(  Stephen D. Tice  ) -- S -- (   first appearances are more deceitful   )
 \                 /     |     \  than in politics.     - David Hume -  /