Mersenne Digest Sunday, August 6 2000 Volume 01 : Number 766
----------------------------------------------------------------------
Date: Thu, 3 Aug 2000 22:54:39 -0700 (PDT)
From: [EMAIL PROTECTED] (Gordon Irlam)
Subject: Re: Mersenne: Undecipherable And Bounced Transmission - probable SPAM!
> well, it had forged headers similar to spam....
Yes. It was spam. My apologies.
The annoying thing is it was sent to [EMAIL PROTECTED]
rather than [EMAIL PROTECTED] [EMAIL PROTECTED] is supposed
to be a secret address that nobody knows about that is the final
address majordomo forwards a message to after it has been accepted
at [EMAIL PROTECTED] and has made it past the anti-spam features
in majordomo designed to block things like this.
Unfortunately it looks like sendmail is leaking information in
the headers it sends out that makes it possible for people to
discover the supposedly secret [EMAIL PROTECTED] address.
I've changed things so that the spammers won't be able to use
this address for spamming the list any longer. Unfortunately
I don't know how to stop the new outgoing address name from
leaking out in the headers so that a smart spammer who finds
an archived copy of a list message on the web might once again
be able to figure out how to spam the list.
I am open to suggestions.
I use sendmail and majordomo, and my sendmail aliases file looks
something like:
mersenne: "|/home/gordoni/local/linux/majordomo/wrapper resend -l mersenne
mersenne-outgoing-queue99"
mersenne-digest: mersenne
mersenne-outgoing-queue99:
:include:/home/gordoni/local/linux/majordomo-data/lists/mersenne,
"|/home/gordoni/local/linux/majordomo/wrapper digest -r -C -l mersenne-digest
mersenne-digest-outgoing-queue99"
mersenne-digest-outgoing-queue99:
:include:/home/gordoni/local/linux/majordomo-data/lists/mersenne-digest
Although the queues aren't actually called "-outgoing-queue99".
thanks,
gordoni (list admin)
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt
------------------------------
Date: Fri, 4 Aug 2000 00:59:42 -0500
From: "Levi Broderick" <[EMAIL PROTECTED]>
Subject: Re: Mersenne: Undecipherable And Bounced Transmission
I, personally, think that it is an advertisement for a computer system. The
phrases 'Com port', 'USB port', 'Email', and 'Excel', among others, are
prominent in the message.
Just my two cents,
~ Levi :o)
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt
------------------------------
Date: Fri, 04 Aug 2000 07:45:09 -0700
From: Spike Jones <[EMAIL PROTECTED]>
Subject: Mersenne: new prefixes...
GIMPS allows us to use some unfamiliar prefixes. For instance, we
are much cheered by our recent landmark of exceeding 1 trillion
operations per second, 1 teraflops. Consider the total floating point
operations performed by GIMPS. We finished our first petaflop
(quadrillion floating point operations) within a few days of its
inception,
and finished our first quintillion floating point operations (exaflop)
in
May of 1998.
The milestone we cross this week is 50 quintillion floating
point operations, or 50 exaflop. Growing at our current
rate, we should get our next prefix, the zettaflop, or sextillion
operations per second (ya gotta love the sound of that, sextillion)
around November 2007.
Then, the next prefix, for 1 septillion flops or yottaflop, would
occur sometime in 2016, with some pretty optimistic growth
assumptions perhaps. {8-] spike
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt
------------------------------
Date: Fri, 04 Aug 2000 09:11:08 -0700
From: Stefan Struiker <[EMAIL PROTECTED]>
Subject: Mersenne: Likelihood Of Small Factors
Question:
>From a number theoretic point of view, does the
likelihood of small (52 through 59 bits) factors increase
with exponent size?
Regards,
Stefanovic
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt
------------------------------
Date: Fri, 4 Aug 2000 09:32:14 -0700
From: Paul Leyland <[EMAIL PROTECTED]>
Subject: RE: Mersenne: Likelihood Of Small Factors
No, it decreases, and eventually becomes identically zero.
All prime factors of M(p) must be of the form 2kp+1. Once p reaches 2^58 we
can guarantee that there are no factors of this form which are less than 59
bits!
Paul
> -----Original Message-----
> From: Stefan Struiker [mailto:[EMAIL PROTECTED]]
> Sent: 04 August 2000 17:11
> To: [EMAIL PROTECTED]
> Subject: Mersenne: Likelihood Of Small Factors
>
>
> Question:
>
> From a number theoretic point of view, does the
> likelihood of small (52 through 59 bits) factors increase
> with exponent size?
>
> Regards,
> Stefanovic
>
> ______________________________________________________________
> ___________
> Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
> Mersenne Prime FAQ --
> http://www.exu.ilstu.edu/mersenne/faq-mers.txt
>
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt
------------------------------
Date: Fri, 4 Aug 2000 19:22:46 +0200 (MET DST)
From: [EMAIL PROTECTED]
Subject: Re: Mersenne: Likelihood Of Small Factors
If you are trying to factor Mp = 2^p - 1 where p is prime and
p > 2^58, then Mp can have no 59-bit factors
[All factors are == 1 (mod 2*p).]
But when factoring a random integer n, we estimate the probability
of a factor between 2^51 and 2^59 as
1 - product (1 - 1/q)
2^51 < q < 2^59
q prime
That is, we model this with a separate event for each
potential prime divisor. You can approximate this
probability by first approximating the logarithm of the product
by an integral. The probability does not depend on n.
If we want to restrict the product to primes q == 1 (mod p),
as when factoring Mp, then (1 - 1/q) becomes (1 - p/(q-1)),
the probability that 2 is a ((q-1)/p)-th residue mod q
This is approximately (1 - 1/q)^p.
There are only 1/(p-1) times as many potential choices for q,
but each such q has approximately p times as much impact.
To a first approximation, the success probability remains
stable (i.e., reaches a limit other than 0 or 1)
as the exponent increases.
Peter Montgomery
- ------------
Question:
>From a number theoretic point of view, does the
likelihood of small (52 through 59 bits) factors increase
with exponent size?
Regards,
Stefanovic
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt
------------------------------
Date: Fri, 4 Aug 2000 13:20:36 -0500
From: "Griffith, Shaun" <[EMAIL PROTECTED]>
Subject: Mersenne: Pinky and the Brain
Since the list seems to be a bit slow, I thought I would throw this out for
your amusement:
http://www.tveyes.com/database/expand.asp?ln=2001719&Key=prime%20number
<http://www.tveyes.com/database/expand.asp?ln=2001719&Key=prime%20number>
[adjusted for punctuation, etc.]
YTV 8/4/00 - 12:49 PM...
[Brain] Dr. Bill Hubbard of the Institute for Advanced Studies was computing
algorithms to the power of 27,000 when the cosine button of his calculator
became stuck, causing his results to increase exponentially. Unaware of the
problem, he continued to use the Briggsian system. Needless to say, the sine
curves of his indices were distorting along a quadratic progression, and
what he saw when he looked down filled him with an unspeakable terror, for
it was the largest prime number ever computed!
[Pinky] [yawns]
BTW, I found this site amusing if not exactly useful...
- -Shaun
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~
Quantum Mechanics: The dreams stuff is made of
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt
------------------------------
Date: Sat, 5 Aug 2000 14:09:16 +0800
From: [EMAIL PROTECTED]
Subject: Re: Mersenne: Undecipherable And Bounced Transmission
- --0__=4rNF7RkrA57WUpKch3KMB5kt8yYjcjoCEIowQoqcyhwsDLYZWBYSQpKD
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
It was not "undecipherable". It was encoded in the
Chinese Big-5 code (used in Taiwan).
The message is to solicit interested party to attend
a class on how to use FAX with computer as the first step
to e-commerce.
It has a contact number of 02-25705975 for enrollment.
HweeBoon
- ---------------------- Forwarded by Low Hwee Boon/KRDL on 08/05/2000 02:06 PM
- ---------------------------
Stefan Struiker <[EMAIL PROTECTED]> on 08/04/2000 01:25:49 AM
To: "Robert G. Wilson v" <[EMAIL PROTECTED]>
cc: (bcc: Low Hwee Boon/KRDL)
Subject: Re: Mersenne: Undecipherable And Bounced Transmission
- --0__=4rNF7RkrA57WUpKch3KMB5kt8yYjcjoCEIowQoqcyhwsDLYZWBYSQpKD
Content-type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-transfer-encoding: quoted-printable
"Robert G. Wilson v" wrote:
> Et al,
>
> I did and deleted it. Do you need it for any reason? Bob.
>
> Stefan Struiker wrote:
>
> > To All:
> >
> > Got a strange (Japanese or Russian?) email from
> >
> > [EMAIL PROTECTED]
No, but my curiosity is piqued [1] because there are legible
references to "486DX-100" and othersuch, which make it quite
possibly a Farsi, say, communiqu=E9 to the Mersenne Ring by someone
in that region, and [2] because every attempt to reply to the sender an=
d
to others listed came back with errors. If Farsi is right to left, I'v=
e struck
out,
but Turkish, perhaps?
Always The Detective,
Stefanovic
_______________________________________________________________________=
__
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.exu.ilstu.edu/mersenne/faq-mers.t=
xt
=
- --0__=4rNF7RkrA57WUpKch3KMB5kt8yYjcjoCEIowQoqcyhwsDLYZWBYSQpKD--
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt
------------------------------
Date: Sun, 6 Aug 2000 19:00:43 +0200
From: "Steinar H. Gunderson" <[EMAIL PROTECTED]>
Subject: Re: ot: p3 motherboards (was Re: Mersenne: Date: Sun, 30 Jul 2000 18:53:50
+0200)
On Tue, Aug 01, 2000 at 05:39:04PM +0200, Sylvain PEREZ wrote:
>What a strange idea you have!
This absolutely isn't strange -- running a torture test doesn't only stress the
hardware by doing FFTs, it also _checks_ the results :-) Great for stability
testing (but also try a program like VNC that stresses the network and graphics
controller quite well -- Prime95/mprime only checks the CPU, RAM and
motherboard).
/* Steinar */
- --
Homepage: http://members.xoom.com/sneeze/
_________________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.exu.ilstu.edu/mersenne/faq-mers.txt
------------------------------
End of Mersenne Digest V1 #766
******************************
