yoyo@home

Message
Author
User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

yoyo@home

#1 Post by rebirther » 14.07.2007, 22:17

http://yoyo8.homedns.org:88/yoyo/

Projektbetreiber ist yoyo vom Team Rechenkraft.net

Bereich Mathematik demnächst...

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#2 Post by rebirther » 10.08.2007, 21:40

Neue Adresse für den Hauptserver:
http://www.rechenkraft.net/yoyo/

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#3 Post by rebirther » 21.11.2007, 09:41

Nach dem Golomb Lineal gibt es jetzt wieder etwas neues, vorab als Test nur in den Projekteinstellungen im Account zu finden:

evolution@home

User avatar
yoyo
Project Admin
Posts: 16
Joined: 01.07.2007, 18:41

#4 Post by yoyo » 21.11.2007, 14:18

... und natürlich in unserem Forum.
yoyo

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#5 Post by rebirther » 06.01.2008, 18:50

06 January 2008
1.04 of evolution@home
Version 1.04 will no more crash if the work unit restarts. But it starts from the very beginning, it does not resume the work unit.
BOINC Neustarts sollten jetzt kein Problem mehr darstellen.

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#6 Post by rebirther » 08.03.2008, 23:09

neues Unterprojekt im Test: Muon

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#7 Post by rebirther » 05.04.2008, 18:10

Neu in Muon 1.07:

- number of simulations reduced from 4 to 2, in avarage the wu should now need half of the time.
- muon priority changed from low to background
- displayed used cpu time is now also stored and should not fall back on a restart (at least not so much)
- some files are moved from wu to application. Wus should now be smaller and the application bigger.
- muonfiles.zip is no more sticky for wus which are now generated, so the files is automatic deleted on client side.

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#8 Post by rebirther » 17.08.2008, 21:51

Hello,
be prepared for the "final push for ogr challenge".
This race starts at 1. September 0:00 UTC and ends at 15. September 0:00 UTC.
The team and user with the most credits in this period is the winner.

The goal of this race is to push the OGR project a bit to its final goal. Currently more than 90% are finished and we hope that the project will still finish in 2008. We encourage each participant to enable only the cruncher application during the race and switch muon and evolution@home off.

The OGR project searches for a optimal Golomb rulers of order 25.

Happy Crunching,
yoyo

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#9 Post by rebirther » 26.10.2008, 15:20

2008-10-25: yoyo@home - OGR-25: Successfully Finished
The 8-year-old OGR-25 distributed computing project has just proven conclusively that the predicted shortest 25-mark Golomb ruler is optimal. 'The total length of the ruler is 480, with marks at positions: 0 12 29 39 72 91 146 157 160 161 166 191 207 214 258 290 316 354 372 394 396 431 459 467 480. (This ruler may alternatively be expressed in terms of the distance between those positions, which is how dnetc displays them: 12-17-10-33-19-...).' 124,387 people participated in the project and two people found the shortest ruler, one on October 10, 2007 and the other on March 24, 2008.Yoyo@home user participated on this contest for


Nun offiziell, OGR-25 erfolgreich beendet.

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#10 Post by rebirther » 06.11.2008, 17:55

Das neue 3. Projekt ist nun bekannt: OGR26, erste Tests laufen schon, um es auf BOINC lauffähig zu machen, Stats gibts hier bei dnet schon in Progress.

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#11 Post by rebirther » 07.11.2008, 15:00

07 November 2008 OGR-NG: new beta apps
I added beta applications for OGR-NG (ogr next generation). You must enable test applications to get them.

The effective search rate is actually more efficient as these new applications are using an improved algorithm. The new algorithm is named FLEGE (Feiri-Levet Enhanced GARSP Engine). In technical terms, the number of elements in the 'choose lookup table' has been increased from 48K to 2M elements. This optimization will significantly reduce the number of nodes that we have to search through, sometimes a node improvement of ten-fold or more.

Distributed.net estimates that OGR-26 will take much less time to complete than OGR-25 and will probably be more similar to OGR-24, in terms of computational effort. Beyond that, distributed.net is looking forward to OGR-27 and OGR-28 because the current solutions to at least one of the two is very likely not optimal. OGR-26 is on the way and must be done first since higher-order rulers depend on the proven optimality of previous ones.

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#12 Post by rebirther » 17.12.2008, 16:30

Neues Unterprojekt in Vortests:

Elliptic Curve Method for Integer Factorization (ECM)

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#13 Post by rebirther » 18.01.2009, 06:43

OddPerfect has many tasks that we work on from time to time. The variety of easy and hard tasks keeps things interesting. Yoyo@home's BOINC ecm has already pushed several tasks beyond where I expected to get in all of 2009. In addition we've had the fun of being major contributors to Paul Zimmermann's the Top 10 Factors of 2009.

In the last couple of weeks Yoyo@home's work has made large progress on an easy task, a hard task, and a very hard task. Most work on OddPerfect numbers depends on factor chains. Factor chains use the property that if a number "x" divides an odd perfect number, then the factors of a related number y also divide the odd perfect number. These factors of y can then be used to generate still more factors - so the more factors we can find the better we can build factor chains.

The easy task is finishing partial factorization that have 120-160 digits left. The hard task is factoring numbers 135-500 digits with no known factors. The very hard task is factoring the five "most wanted" numbers.

In the easy task we extended previous work on some easy numbers. The life cycle of these numbers is several stages of searching for numbers with ecm and then turning them over to factoring by the Number Field Sieve (NFS). We pushed one set of these numbers to the next stage and found some factors in the process. In the future we will give these numbers one more stage then turn them over to NFS. There is another batch of over 100 easy numbers queued up for the first stage work, and several more batches that will be queued in the future.

In the hard task we found a lot of "first known factors." Most of our records in the top 10 factors of 2009 come from this task. We should have found about 1/3 of the 50-digit factors in these numbers. In the future we will look for 50 and 55 digit factors of these numbers.

In the very hard task yoyo@home is expected to find 1/3 of the 55 digit factors in the five "most wanted" numbers. We have finished two and are 90% done with the third - no factors found yet. We will keep returning to these numbers.

I have a couple of other tasks queued up and I have plans for still more tasks in the future. I'll tell you about these tasks as we get to them in the queue.

Thanks to all the BOINC crunchers for the great progress in these tasks.

William

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#14 Post by rebirther » 23.02.2009, 20:29

23 February 2009 ECM: OddPerfect
This weekend yoyo@home finished work on a batch of small composites and advanced a special project for Vanishing Fermat Quotients. We factored a dozen of the small composites. The most fun was 158 digit number for which we found two 40 digit factors. Only about 6 in 10,000 composites have two such factors, and even then we often miss one of the factors until later.

Vanishing Fermat Quotients (VFQ) happen when, for primes p and q, q2 divides pq-1. Pace Nielsen has many cases are simplified when the VFQs have two large factors, and the simplest way to verify that is to show two large factors. yoyo@home has found four new VFQ factors so far.

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#15 Post by rebirther » 24.02.2009, 19:10

24 February 2009 OGR: OGR-26 finished, -27 started
OGR-26 is now officially completed. I cancelled all remaining OGR-26 wus. Wus which are now sent are OGR-27 wus.
Due to variations in complexity, distributed.net expects that OGR-27 will take significantly longer than OGR-26 did. It is difficult to provide a precise estimate but one extremely rough guess is about 7 years, assuming no increase in computing power and that the size estimation sampling reflects the entire stubspace.
There is one thing that is different with OGR-27 than with previous OGR projects: distributed.net is confident that we will discover a better ruler for OGR-27 than the one which is known to be optimal currently.

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#16 Post by rebirther » 07.05.2009, 17:08

Na gucke, da hab ich was gefunden, für die paar Futzeln schon ne gute Leistung :gamer1:
2009-05-07
17:35:48
Rebirther
OddPerfect 521_103M.C202
(B1 = 11000000) 246103340516858910456236224045637567
36

User avatar
Saenger
Guest
Posts: 7
Joined: 11.04.2008, 16:59
Location: Bremen

#17 Post by Saenger » 28.11.2009, 18:48

Yoyo hat jetzt die Windoof-Anwendung mit Wine irgendwie so hingebastelt, das es auch geht.
Er hat's hier beschrieben:

1) So noch nicht geschehen, Konto bei Yoyo@Home erstellen
2) Die Anwendung (Hier als gepacktes Archiv erhältlich) herunterladen und den Inhalt in projects/www.rechenkkraft.net_yoyo entpacken
3) BOINC neu starten
4) Eine muon-Wutz herunterladen.

Ach ja, natürlich nur unter folgender Voraussetzung:
Es muss auf dem betreffenden Rechner natürlich wine installiert sein, und zwar unter usr/bin/wine

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#18 Post by rebirther » 31.01.2010, 17:39

Es werden gerade RepUnit ECM WUs getetest, dies erweitert die ECM Reihe um ein weiteres Unterprojekt, kann man leider aber nicht selbst auswählen.

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#19 Post by rebirther » 24.04.2010, 22:57

neues Unterprojekt noch mit kurzer Deadline:

Euler

User avatar
rebirther
Admin
Posts: 2902
Joined: 19.12.2005, 00:59
Location: Germany

#20 Post by rebirther » 13.06.2010, 06:51

12 June 2010 Euler625 new search
A new Euler application is now on test for Linx32 and Linux64 systems. To get the app you must enable test applications. Workunits are sometimes created for testing. The new Euler application searches now up to 250.000. There are 510.000 workunits in total. But first we need testers.
Euler wird erweitert, leider dauern die WUs jetzt um das vielfache länger, die Deadline ist etwas zu kurz gehalten, erste Tests auf Linux, nach erfolgreichen Lauf werden 510.000 WUs ausgegeben.

Post Reply