Google Groups unterstützt keine neuen Usenet-Beiträge oder ‑Abos mehr. Bisherige Inhalte sind weiterhin sichtbar.

Fwd: Warhead maims innocent Ethernet! Details at 11

17 Aufrufe
Direkt zur ersten ungelesenen Nachricht

John Gardiner Myers

ungelesen,
16.03.1990, 12:11:2216.03.90
an
--- Start of forwarded messages ---

Date: Wed, 14 Mar 90 20:49 EST
From: Steve Strassmann <st...@media-lab.media.mit.edu>
To: unix-...@ai.ai.mit.edu
Subject: [dre...@decwrl.dec.com: what to do with a PDP-11/73]

Date: Wed, 14 Mar 90 16:50:22 -0800

From: dre...@decwrl.dec.com (Raymond Drewry)
From: Michel Jackson <jac...@shs.ohio-state.edu>
From: ka...@giza.cis.ohio-state.edu (Karl Kleinpaste)
Newsgroups: osu.hacks
Subject: How do you measure nuclear warhead yield?

This is something I learned at the USENIX Conference in January that
I've been meaning to post here, but have managed to forget about until
now.

While chatting with some network acquaintances at the hotel bar (all
the important discussion occurs at the bar, of course, preferably well
past midnight), a friend who does sysadmin work at Los Alamos National
Labs told us a marvelously funny story about how the fun folks at LANL
measure yield from nuclear detonations. After all, they have to
experiment, I guess, and one has to learn how much bang-for-the-Mbuck
one is getting.

The solution at LANL (note that this is now an 8-week-old memory,
details may be somewhat inaccurate):

Find a Qbus-based PDP-11 (e.g., 11/73) "which you no longer love."
Install a DEQNA ethernet controller card in the backplane. Park the
box at/near/over the hole. Connect a cable to the DEQNA and drop it
down into the hole.

DEQNAs have a TDR (time domain reflectometer) built right into the
controller. TDR is useful for finding cable shorts and, in general,
learning the length of one's ethernet cable.

Before detonation, begin having the PDP-11 repeatedly exercise the
DEQNA's TDR, recording and transmitting the length determined to some
other (presumably distant :-) site.

Detonate. As the beastie blows things to smithereens all around
itself, the cable will be rapidly eaten away. TDR readings from the
DEQNA will show a drastically reducing cable length. The speed with
which the cable, ah, degenerates will correlate very closely with
warhead yield.

Just think, your tax dollars at work, ridding the world of PDP-11s...

--karl

PS- No, I'm not kidding. Not a word of it.


--- End of forwarded messages ---

Barry Shein

ungelesen,
16.03.1990, 22:29:2716.03.90
an

Re: nuking DEQNA's

I doubt very much this is true.

I've heard similar stories dating back to the early 60's, mostly
involving video signals racing the vaporization ring down the wire.

Then again maybe everything is true, perhaps you can trade in those
vaporized video cameras for a vaporized DEQNA in some DEC upgrade
program I haven't heard about.
--
-Barry Shein

Software Tool & Die | {xylogics,uunet}!world!bzs | b...@world.std.com
Purveyors to the Trade | Voice: 617-739-0202 | Login: 617-739-WRLD

Steven M. Bellovin

ungelesen,
17.03.1990, 21:54:2617.03.90
an
In article <1990Mar17....@world.std.com>, b...@world.std.com (Barry Shein) writes:
>
> Re: nuking DEQNA's
>
> I doubt very much this is true.
>
> I've heard similar stories dating back to the early 60's, mostly
> involving video signals racing the vaporization ring down the wire.

I can't say whether or not this particular story is true; however, it
is quite definitely true that measuring the collapse of a piece of coax
is felt to be a reliable way to measure the yield of a nuclear weapon.
The technique became a point of contention between the U.S. and the U.S.S.R.
a few years ago, during negotiations about ratifying a test ban treaty.
This particular treaty limits underground tests to 150 kilotons; although
not yet ratified, both sides have been observing it. Verification is the
sticking point. The U.S.S.R. prefers reliance on seismic techniques,
validated in this case by on-site baseline measurements of the geological
characteristics of the test zones. Thus, non-intrusive verification of
actual nuke tests becomes possible. The U.S. position -- under an
administration that wasn't particularly friendly towards the whole
concept -- was that seismic methods were not reliable enough, and that
on-site verification was necessary. They suggested the crushed-coax
technique -- called CORRTEX, I think, or something similar -- as a
method that was both reliable and also sufficiently simple as to allay
fears that other intelligence was being gathered about the blast. Matters
went sufficiently far that it was demonstrated to the Soviets, though
I don't recall whether it was done in this country or in the Soviet Union.
Nor do I know if Reagan was willing to sign blanket export licenses for
PDP-11s with DEQNAs...

0 neue Nachrichten