Mersenne Digest Sunday, 10 May 1998 Volume 01 : Number 357 ---------------------------------------------------------------------- From: Paul Leyland Date: Thu, 7 May 1998 05:25:23 -0700 Subject: Mersenne: M(3547) co-factor is a 1039-digit prime. The co-factor of M(3547) was given as a strong pseudoprime of 1039 digits. I had not seen a proof of its primality, so I've just spent 23.6 hours computation on a PII-300 to find a proof with Morain's ECPP program. Anyone who *really* wants the output from ECPP is welcome to ask for an emailed copy, but please be aware that it runs to very nearly a megabyte. Paul ------------------------------ From: John Simmons Date: Thu, 07 May 1998 12:03:10 -0700 Subject: Re: Mersenne: Prime95 version 16.2 At 12:05 PM 5/5/98 -0400, George Woltman wrote: >Hi all, > > A typo caused version 16.1 to underestimate factoring time >by a factor of 10. The fixed executable is now available at >http://www.magicnet.net/~woltman/prime95.zip > I tried running this version and was getting Error 87 messages when trying to contact the server. Went back to 15.4. /*===================================================*/ There are three types of lies... lies ("You'll never need more than 64K"), damn lies ("Windows is more stable than Linux"), and press releases ("It will be released as scheduled"). John Simmons (Redneck Techno-Biker) jms1@home.com http://www.members.home.net/jms1/index.html /*===================================================*/ ------------------------------ From: "Chuck W. " Date: Thu, 7 May 1998 16:49:19 -0700 (PDT) Subject: Re: Mersenne: Prime95 version 16.2 Did you remember to re-name the dll files? On Thu, 7 May 1998, John Simmons wrote: > At 12:05 PM 5/5/98 -0400, George Woltman wrote: > >Hi all, > > > > A typo caused version 16.1 to underestimate factoring time > >by a factor of 10. The fixed executable is now available at > >http://www.magicnet.net/~woltman/prime95.zip > > > > I tried running this version and was getting Error 87 messages when trying > to contact the server. Went back to 15.4. > > > /*===================================================*/ > There are three types of lies... > lies ("You'll never need more than 64K"), > damn lies ("Windows is more stable than Linux"), and > press releases ("It will be released as scheduled"). > > John Simmons (Redneck Techno-Biker) > jms1@home.com > http://www.members.home.net/jms1/index.html > /*===================================================*/ > > - - I tell you the truth, we speak of what we know, and we testify to what we have seen, but still you people do not accept our testimony. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ : WWW+PGP: http://www.silverlink.net/poke : : E-Mail: chuckw@silverlink.net : ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ : According to Section 227(b)((3)(B) of US Code Title 47 I am entitled : : to $500 per un-solicited commercial e-mail. : ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ------------------------------ From: John Simmons Date: Thu, 07 May 1998 16:58:09 -0700 Subject: Re: Mersenne: Prime95 version 16.2 At 04:49 PM 5/7/98 -0700, Chuck W. wrote: > > >Did you remember to re-name the dll files? > >On Thu, 7 May 1998, John Simmons wrote: > >> I tried running 16.2 and was getting Error 87 messages when trying >> to contact the server. Went back to 15.4. >> >> ummmmm, huh? I guess I'll go re-read the docs again. Maybe I missed that part... /*===================================================*/ There are three types of lies... lies ("You'll never need more than 64K"), damn lies ("Windows is more stable than Linux"), and press releases ("It will be released as scheduled"). John Simmons (Redneck Techno-Biker) jms1@home.com http://www.members.home.net/jms1/index.html /*===================================================*/ ------------------------------ From: "John R Pierce" Date: Thu, 7 May 1998 18:10:40 -0700 Subject: Re: Mersenne: Prime95 version 16.2 > I tried running 16.2 and was getting Error 87 messages when trying > to contact the server. Went back to 15.4. I got some of those too, but it turned out the entropia server was down temporarily, I couldn't hit the website either.... Also, my firewalled systems, I had to not only rename the httpnet.dll to primenet.dll but I also had to edit primenet.ini and reenter the password (removing the proxymask=1 so that it would reencode it), seemingly, the proxy password encryption has changed or something? - -jrp ------------------------------ From: David Underbakke Date: Thu, 07 May 1998 21:18:43 -0500 Subject: Re: Mersenne: Prime95 version 16.2 At 12:03 PM 5/7/98 -0700, you wrote: >At 12:05 PM 5/5/98 -0400, George Woltman wrote: >>Hi all, >> >> A typo caused version 16.1 to underestimate factoring time >>by a factor of 10. The fixed executable is now available at >>http://www.magicnet.net/~woltman/prime95.zip >> > >I tried running this version and was getting Error 87 messages when trying >to contact the server. Went back to 15.4. > > I have been using 16.1 and 16.2 with no problems at all. I did have one initial problem with 16.1, I initially only copied the prime95.exe file, and not all the files in the zip file (like primenet.dll, which is newer in the new release). After copying all the files, everything works fine. Lastly, a) make sure you do not have any stuff queued in prime.spl before updating, and b) if you are using the proxy version httpnet.dll, rename it to primenet.dll before usage. ______________________________________________________ _/_/_/ David Underbakke david@underbakke.com _/_/_/ _/_/_/ Aramit Technologies david@aramit.com _/_/_/ _/_/_/ Voice:(612) 646-3401 Fax:(612) 603-9763 _/_/_/ _/_/_/ Micrographic Retrieval Delivery Solutions _/_/_/ _/_/_/ Windows NT Novell Lantastic DEC Internet _/_/_/ ------------------------------ From: JGrif91665 Date: Fri, 8 May 1998 15:25:34 EDT Subject: Mersenne: Prime95 concerns The local college where I have been running Prime95 now has some concerns about running the program on their computers. It may help if members that are running Prime95 on multiple computers, in a college or at work, share we them the reasons for participating in GIMPS, any problems that the program causes or other information they may need to ease their concerns about Prime95 would be appreciated. Send e-mail to: Ginger Dennis: gdinnis@ntcc.cc.tx.us Patsy Armstrong: parmstrong@ntcc.cc.tx.us Thank you, James Griffin ------------------------------ From: SpoolDog Date: Sun, 10 May 1998 00:49:35 EDT Subject: Mersenne: program Hey if some one sends me the algorithm and some basic flow structure guidelines for the things used in prime95 will try and create new faster program. dont expect any results any time soon. will be using different language this is just something for me to try but i dont have the info(algorithms, etc). ------------------------------ From: "John R Pierce" Date: Sat, 9 May 1998 23:19:52 -0700 Subject: Re: Mersenne: program SpoolDog writes... >Hey if some one sends me the algorithm and some basic flow structure >guidelines for the things used in prime95 The full source is available on the mersenne.org website. >will try and create new faster program. >dont expect any results any time soon. will be using different language Its heavily hand tuned heavily pipelined pentium floating point assembler language. The "C" code in Prime95 is purely user interface, setup, etc. The entire core is assembler. >this is just something for me to try but i dont have the info(algorithms, >etc). The core algorithm is simple. massive large Fast Fourier Transforms performing the multiplications that are the core of implementing the Lucas Lehmer test. good luck! ------------------------------ End of Mersenne Digest V1 #357 ******************************