The Mersenne Newsletter, issue #14 May 22, 1998 IMPORTANT! Version 16 of the program released --------------------------------------------- We are running out of exponents to test below 5,260,000!! There is less than a two week supply of exponents to assign. Please download version 16 of the program, it can test exponents as high as 20,500,000. If you are running version 14 of the program, you can wait until your current range completes before upgrading. If you are upgrading from version 15, follow these simple steps: 1) Download the new version. 2) If the program needs to contact the server (a prime.spl file is present in the directory), then let the old version report these spooled results. 3) Install the new version on top of the old version. (If you renamed HttpNet.dll to PrimeNet.dll for version 15, then rename the file again for version 16. If you use a proxy server userid/password in the primenet.ini file, you must set ProxyMask=0 and re-enter the password in the file.) 4) Start the program. Optionally read the whatsnew.txt file to see what new features are available. GIMPS is now almost entirely automated on the PrimeNet server. You can watch the hourly progress of the search on the server's status page at http://entropia.com/primenet/status.shtml. News on distributed computing aspects of automating GIMPS, new software availability and related information is also there. Manual testing "ranges" should be acquired from PrimeNet. Click the link at the top of the status page, and use the simple web forms to get exponents and check in your test results. The forms accept results from Prime95/NTPrime (versions 16.3, 15.4, 14.4), MacLucas V1.4, MPrime V15.4 and V16.3, and PRIMEOS2 V3.0 programs. The most frequently asked questions and answers about using Prime95 and NTPrime with PrimeNet are also linked at the top of the status page. If your question is not covered, network problems or PrimeNet account issues can be sent to Scott Kurowski, primenet@entropia.com. The PrimeNet status page highlights how quickly exponents below 5,260,000 are disappearing. Fortunately, version 16 should keep our computers busy for years to come! Status for exponents below 2,655,000 ------------------------------------ Since the last newsletter on February 2, the number of untested exponents has dropped from 657 to 116, with only a 1% chance of finding a new Mersenne prime. Status for exponents from 2,655,000 to 5,260,000 ------------------------------------------------ Since February 2, we've proved 20,370 Mersenne numbers composite. There are now 24,688 exponents left to test. We've reduced the work effort by 1,752 CPU years, leaving an estimated 2,555 P-90 CPU years to complete. IMPORTANT REMINDER: Policy on Manually Reserved Ranges ------------------------------------------------------ Ranges may be returned to the available pool if no results are reported during a 4 month period. Best wishes and good luck. Maybe you will be the one to find the 38th Mersenne prime! George Woltman woltman@magicnet.net