Mersenne Digest Monday, December 6 1999 Volume 01 : Number 668
----------------------------------------------------------------------
Date: Sat, 4 Dec 1999 08:17:17 +0000
From: "Steinar H. Gunderson" <[EMAIL PROTECTED]>
Subject: Mersenne: Re: Error 2250 using Linux mprime
On Thu, Dec 02, 1999 at 03:45:30PM -0600, Willmore, David wrote:
>Strangely, I found a libnss_dns-2.0.7.so and a symlink pointing to it
>called, of all things, libnss_dns.so.1, not ".2" like I would have expected.
To me, this sounds like a new interface in glibc 2.1.
>So, I created a ".2" symlink and, what do you know, it works.
If it works, you're lucky. Most likely something is broken, but perhaps
mprime doesn't use it :-)
Strange thing EntropiaIP= didn't work for you, though. I'd still say that
the best way to do this, is making an sprime binary with static NSS compiled
in. I think that would solve most of the problems we have with older glibc-
versions. Sure, we would lose some of the advantages of NSS (namely the
ability to use whatever method instead of resolv.conf), but we could always
leave that in in mprime.
/* Steinar */
- --
Homepage: http://members.xoom.com/sneeze/
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
------------------------------
Date: Sat, 4 Dec 1999 17:15:57 +0000
From: "Steinar H. Gunderson" <[EMAIL PROTECTED]>
Subject: Mersenne: Re: Strange Prime95 behavior. Can anyone explain ???
On Sat, Dec 04, 1999 at 08:55:02AM -0500, Olivier Langlois wrote:
>Yesterday, my Prime95 client have contacted the Primenet server to get more
>work to do and since then, my client stopped to do the LL test on the
>unfinished prime M7943231.
>From what I can see, your computer has started factoring another number.
This is completely normal (see the README file). Once the factoring is done,
Prime95 should continue your LL-test.
The reason for this behaviour is, that if it was to complete the LL test,
and then quickly find a factor for the next number, you would suddenly be
left with no work left :-)
>33233791 * 32 0.5 1437 64.5 04-Dec-99
>01:55 358BeaubienE
This seems to be the number. Note that it's only factored to 32 bits, and
probably will be factored to 67 bits (just a guess, there are tables for
this) before the LL test continues.
/* Steinar */
- --
Homepage: http://members.xoom.com/sneeze/
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
------------------------------
Date: Sat, 04 Dec 1999 13:18:46 -0600
From: Ken Kriesel <[EMAIL PROTECTED]>
Subject: Re: Mersenne: Re: Strange Prime95 behavior. Can anyone explain ???
At 05:15 PM 1999/12/04 +0000, "Steinar H. Gunderson"
<[EMAIL PROTECTED]> wrote:
>On Sat, Dec 04, 1999 at 08:55:02AM -0500, Olivier Langlois wrote:
>>Yesterday, my Prime95 client have contacted the Primenet server to get more
>>work to do and since then, my client stopped to do the LL test on the
>>unfinished prime M7943231.
>
>From what I can see, your computer has started factoring another number.
>This is completely normal (see the README file). Once the factoring is done,
>Prime95 should continue your LL-test.
>
>The reason for this behaviour is, that if it was to complete the LL test,
>and then quickly find a factor for the next number, you would suddenly be
>left with no work left :-)
>
>>33233791 * 32 0.5 1437 64.5 04-Dec-99
>>01:55 358BeaubienE
>
>This seems to be the number. Note that it's only factored to 32 bits, and
>probably will be factored to 67 bits (just a guess, there are tables for
>this) before the LL test continues.
>
>/* Steinar */
The table below is from a posting by George Woltman some time ago:
Numbers above are factored to
- ------------- ---------------
71000000 2^72
57020000 2^71
44150000 2^70
35100000 2^69
28130000 2^68
21590000 2^67
17850000 2^66
13380000 2^65
8250000 2^64
Trial factoring runtime can be very significant.
For example, M33233791 will take a few weeks on a Pentium-200.
(But, if it finds a factor, that could shrink to mere seconds.)
Ken Kriesel
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
------------------------------
Date: Sun, 05 Dec 1999 06:27:08 +0000
From: Michael Oates <[EMAIL PROTECTED]>
Subject: Re: Mersenne: Strange Prime95 behavior. Can anyone explain ???
Olivier,
Don't worry, when it has finished the factoring it will continue where it
left off with the LL test.
>Hi,
>
>Yesterday, my Prime95 client have contacted the Primenet server to get more
>work to do and since then, my client stopped to do the LL test on the
>unfinished prime M7943231.
>
>Can anyone explain why ?
>I hope there is something I can do to finish to test this prime. I would
>find it a little bit frustrating to have wasted all this time for nothing.
>
>Thank you
Mike,
- --
ATLAS CELESTE - Bevis Star Atlas - & "The CD-ROM"
A very rare atlas found at the Godlee Observatory
http://www.u-net.com/ph/mas/bevis/
Astronomy in the UK http://www.ph.u-net.com
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
------------------------------
Date: Sun, 5 Dec 1999 15:07:46 -0500
From: "Frank_A_L_I_N_Y" <[EMAIL PROTECTED]>
Subject: Mersenne: questions
why is f(0) in an ll test = 4
also when i went to sshool (i'm 34) 7 mod 8 == 7
i have seen 7 mod 8 == -1
is -1 == 7 mod 8
which is more correct
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
------------------------------
Date: Sun, 5 Dec 1999 16:51:07 -0500
From: "Frank_A_L_I_N_Y" <[EMAIL PROTECTED]>
Subject: Mersenne: sorry
cant belive i spelled school sshool
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
------------------------------
Date: Mon, 6 Dec 1999 00:05:07 +0100
From: "Otto Bruggeman" <[EMAIL PROTECTED]>
Subject: Re: Mersenne: sorry
> why is f(0) in an ll test = 4
It is, just accept it...
> also when i went to sshool (i'm 34) 7 mod 8 == 7
> i have seen 7 mod 8 == -1
> is -1 == 7 mod 8
> which is more correct
- -1 + 8 = 7
so 7 mod 8 = 7 or -1 depends on how you look at it...
> cant belive i spelled school sshool
cant believe you spelled believe belive.. ;)
Sorry, couldn't resist...
Otto
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
------------------------------
Date: Sun, 5 Dec 1999 18:18:00 -0500 (EST)
From: Lucas Wiman <[EMAIL PROTECTED]>
Subject: Re: Mersenne: questions
> why is f(0) in an ll test = 4
Well, it needn't be. There are a number of other possibilities, but 4 is the
smallest...
> also when i went to sshool (i'm 34) 7 mod 8 == 7
> i have seen 7 mod 8 == -1
> is -1 == 7 mod 8
> which is more correct
They are both correct. If you have a congruence relation (mod n), then
you can add n to either side (it's like adding or subtracting zero).
- -Lucas
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
------------------------------
Date: Mon, 6 Dec 1999 00:26:26 +0100
From: Preda Mihailescu <[EMAIL PROTECTED]>
Subject: Re: Mersenne: sorry
>
> > why is f(0) in an ll test = 4
>
> It is, just accept it...
I think the Lehmer sequence is L(4,1) and f(n)=v_{2^n}, the should at least be correct
in
an indicative way - you should verify too.
>
> > also when i went to sshool (i'm 34) 7 mod 8 == 7
> > i have seen 7 mod 8 == -1
> > is -1 == 7 mod 8
> > which is more correct
>
> -1 + 8 = 7
>
> so 7 mod 8 = 7 or -1 depends on how you look at it...
>
Or also 7 = 23 mod 8, etc.
They are all literally the same - that's why you speak of an equivalence class.
Now in order to work with equivalence classes you like to chose predilectional
representatives. The most obvious ways to do this for the modulus m are:
A. Choose a representative 0<=a<m
B. Choose a representative -m/2<=a<m/2 (minimize the abolsute value of the
representatives
In your case -1 and 7 are the choices B. and A. for the same equivalence class.
Preda
>
> > cant belive i spelled school sshool
>
> cant believe you spelled believe belive.. ;)
>
> Sorry, couldn't resist...
>
> Otto
>
> _________________________________________________________________
> Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
> Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
------------------------------
Date: Mon, 6 Dec 1999 21:34:11 -0700
From: "Aaron Blosser" <[EMAIL PROTECTED]>
Subject: Mersenne: Total P-90 CPU Year Landmarks
Well, I recently passed the 100 P-90 CPU years landmark. I thought it was
neat. Would've gotten there sooner, but my bro and I lost a few machines to
attrition recently. Sigh...and they were a bunch of dual and quad
PII/PIII's. Double sigh...
I suppose next year we may see Temple-U CAS pass the 1000 mark.
And I must note that none of my LL CPU years belong to US WEST (although I
think some of my 13.236 factoring years do). :-)
It reminded me of how fun it was when I passed the 10 CPU year mark a long
time ago.
So okay, maybe I'm not a math genius. I follow the LL algorithm just fine,
but I like to think that there are lots more like me out there who don't
quite get it all, but have lots of CPU time to donate. I just wanted to let
everyone know that you don't *need* to be a math whiz to contribute to what
I consider a very worthwhile effort.
Just my $0.02 (USD) worth.
Aaron (team "madpoo": currently #14)
_________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers
------------------------------
End of Mersenne Digest V1 #668
******************************
