distributed.net "news"...
Korey Nixon <[email protected]>
| Newsgroups | gmane.comp.distributed-net.rc5 |
|---|---|
| Message-ID | <[email protected]> |
Dear fellow RC5 mailing list members, I was recently reading an article in "Computer Power User" Magazine about "charitable computing" The link to the article online is here: http://www.computerpoweruser.com/editorial/article.asp?article=articles%2Farchive%2Fc0903%2F46c03%2F46c03.asp Here's a quote from the article: *"In the past, researchers often set up volunteer computing programs, attracted a base of users, and then forgot to communicate with the troops. Now that volunteers can switch their CPU time among causes with a single interface, researchers must learn to work in a charity marketplace. 'The fact is there are people behind those computers, and they need reassurance that they aren’t just wasting electricity,' says Anderson. One of the challenges for volunteer computing projects is having the research teams devote time and resources to community cultivation. 'Projects attract users by doing good science and by having Web sites that have up-to-date news and interesting message boards,' he says."* ** Let's face it - the main distributed.net web site is out of date! So I've written some "news" for distributed.net. I've even spiced things up a bit by including a fun contest at the bottom of the news bulletin. *I'm sending this to the mailing list with the best of intentions, because I believe we have an opportunity to attract a significant amount of new users, especially with the work that has been happening with video card (GPU) based clients.* That's why I've taken this step by writing some content for distributed.net's web site. I'm sending this to the mailing list first, in order to receive some feedback. If you think this is a good idea, please let me know. Depending on the feedback received, I may get in touch with the distributed.net administration team to see if they want to use the content. Please find below the content I've written. Thanks, and happy crunching!!! _________________ Distributed.net News: Greetings! It is with pleasure that we announce some new developments here at distributed.net. Projects RC5-72 and OGR-NG are ongoing, with some exciting news for both. *RC5-72:* Current keyrates for RC5-72 are near all time highs. An average of 50 blocks/day has been the norm for some time now, compared to the long term average of 33 blocks/day. These high key rates can be attributed to the launch of several beta clients which run on Graphics Processing Units, or GPUs. On the pre-release client page you can find betas for both NVIDIA and ATI based video cards. These clients are only capable of participating in the RC5-72 project at present. The key rates are very high, as the GPUs of today are much better suited to process RC5-72 work units than CPUs are. We encourage beta testers, so head on over to the pre-release page <link> and download your client today. To use these clients, the following hardware is required at minimum: NVIDIA video cards: Minimum GeForce 8xxx series ATI video cards: Minimum Radeon HD 2xxx series So, just how fast will these clients perform? To use NVIDIA cards as an example: With the entry level (8600) series GeForce cards, you can expect roughly 70 million keys/sec or roughly 1,400 blocks/day. With the higher end cards (GTX 285) you can expect upwards of 500 million keys/sec, or roughly 10,000 blocks/day. ATI cards have similarly impressive numbers. And don't forget - despite RSA security discontinuing the official sponsorship of the RC5-72 project, there still is prize money available! distributed.net has picked up RSA's slack and will provide the winner with US $2000 if a single participant, or US $1000 if on a team, with the other $1,000 going to the winners' team. We encourage you to recruit more CPUs and especially GPUs to participate in the RC5-72 project. Together, we can wrap the project up before we all know it - and hey, you might be $2000 richer for it! *OGR-NG:* Thanks to an impressive initial support base of participants, OGR-27 is already > 1% completed. We are pleased to announce some exciting work being done on new OGR cores, which will provide increased node rates. CPUs that support Streaming SIMD Extensions 2 (SSE 2) instructions (introduced originally in Intel's Pentium 4 processor and still present in many CPUs today) will benefit with faster speeds. See Wikipedia's article ( http://en.wikipedia.org/wiki/SSE2#CPUs_supporting_SSE2 ) on SSE2 to check if your CPU supports SSE2 instructions and will benefit from the new cores. Keep an eye on the pre-release page for release candidates containing the SSE2 optimized cores, and a special thanks to *Craig Johnston* for his coding work. *CONTEST* Many of you participate in distributed.net's projects, but how "hardcore" are you? Maybe you're still crunching RC5-72 on a 386 with no hard drive but have rigged in a network card? Are you? Send us a picture! We're searching for images of the oldest, slowest PC's crunching RC5-72. Send in the system specs along with the image to <user_here>. The winning entry will be posted on our web site. Of course, don't forget to flush in those blocks! End content written. -- Korey Nixon Halifax, NS Canada P: 902-446-5752 E: [email protected] _______________________________________________ rc5 mailing list rc5-Ra3b/QYEcJ3d140v2zMXi0fjHoOT/h/[email protected] http://lists.distributed.net/mailman/listinfo/rc5