| Author |
Message |
|---|
John Honorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
                 
|
|
Welcome to the Year of the Tiger
We are kicking off the 2010 Challenge Series with the Year of the Tiger which just so happens to coincide with Valentine's Day in the West. The Tiger is the third sign in the Chinese Zodiac cycle, and it is a sign of fearlessness. It is considered incredibly brave, evidenced by its willingness to engage in battle. So whether it's your love of Tigers or love of primes, please come join us in an East & West fearless attack on Proth Primes.
A 3 day Challenge is being offered on PrimeGrid's Proth Prime Search (LLR) application. NOTE: Only PPS work will be released during the Challenge. However, PPSE resends will occur. All completed work, PPS or PPSE, will be credited in the Challenge score.
To participate in the Challenge, please select only the Proth Prime Search (LLR) project in your PrimeGrid preferences section. The challenge will begin 14 February 2010 18:00 UTC and end 17 February 2010 18:00 UTC. Application builds are available for Linux 32 bit, Windows 32 bit and MacIntel. These applications will be sent to 64 bit clients. As with all LLR application projects, there is no advantage of 64 bit over 32 bit.
ATTENTION: The primality program LLR is CPU intensive; so, it is vital to have a stable system with good cooling. It does not tolerate "even the slightest of errors." Please see this post for more details on how you can "stress test" your computer. WU's will take ~5 minutes on fast/newer computers and 15+ minutes on slower/older computers. If your computer is highly overclocked, please consider "stress testing" it. Sieving is an excellent alternative for computers that are not able to LLR. :)
Please, please, please make sure your machines are up to the task.
Time zone converter:
The World Clock - Time Zone Converter
NOTE: The countdown clock on the front page uses the host computer time. Therefore, if your computer time is off, so will the countdown clock. For precise timing, use the UTC Time in the data section to the left of the countdown clock.
Scoring Information
Scores will be kept for individuals and teams. Only work units issued AFTER 14 February 2010 18:00 UTC and received BEFORE 17 February 2010 18:00 UTC will be considered for credit. We will use the "prime score" method which is based on the n value (k*b^n-1) to score the challenge. The only difference is that the primary and double checker of a WU will receive the same score.
Therefore, each completed WU will earn a unique score based on its n value. The higher the n, the higher the score. This is different than BOINC cobblestones! A quorum of 2 is NOT needed to award Challenge score - i.e. no double checker. Therefore, each returned result will earn a Challenge score. Please note that if the result is eventually declared invalid, the score will be removed.
For details on how the score is calculated, please see this thread.
At the Conclusion of the Challenge
We kindly ask users "moving on" to ABORT their WU's instead of DETACHING, RESETTING, or PAUSING.
ABORTING WU's allows them to be recycled immediately; thus a much faster "clean up" to the end of an LLR Challenge. DETACHING, RESETTING, and PAUSING WU's causes them to remain in limbo until they EXPIRE. Therefore, we must wait until WU's expire to send them out to be completed.
Please consider either completing what's in the queue or ABORTING them. Thank you. :)
About the Proth Prime Search
The Proth Prime Search is done in collaboration with the Proth Search project. This search looks for primes in the form of k*2^n+1. With the condition 2^n > k, these are often called Proth primes. This project also has the added bonus of possibly finding factors of "classical" Fermat numbers or Generalized Fermat numbers. As this requires PrimeFormGW (PFGW) (a primality-testing program), once PrimeGrid finds a prime, it is then tested on PrimeGrid's servers for divisibility.
Our initial goal was to double check all previous work up to n=500K for odd k<1200 and to fill in any gaps that were missed. We have accomplished that now and have increased it to n=800K. PG LLRNet searched up to n=200,000 and found several missed primes in previously searched ranges. Although primes that small did not make it into the Top 5000 Primes database, the work was still important as it may have led to new factors for "classical" Fermat numbers or Generalized Fermat numbers. While there are many GFN factors, currently there are only about 275 "classical" Fermat number factors known. Current primes found in PPS definitely make it into the Top 5000 Primes database.
Once the 800K goal is reached, we may head to 1M before turning our focus to smaller k values and higher n values. For example, k<300 complete to n=2M, k<600 complete to n=1.5M and so on.
Additional Proth prime testing can be found in PrimeGrid's PRPNet's Proth Prime Search Extended (PPSE) project. This project extends the search range to 1200<k<10000 for n<2M. Current ports available are:
PRPNet Ports
Detailed daily user stats courtesy of Sysadm@Nbg
Today's Prime Finders courtesy of Sysadm@Nbg
port 10000: PPSE n<450K Server : User : Pending Tests
port 11000: PPSE n<450K Server : User : Pending Tests
port 12007: PPSE n>700K Server : User : Pending Tests
For more information about "Proth" primes, please visit these links:
About Proth Search
The Proth Search project was established in 1998 by Ray Ballinger and Wilfrid Keller to coordinate a distributed effort to find Proth primes (primes of the form k*2^n+1) for k < 300. Ray was interested in finding primes while Wilfrid was interested in finding divisors of Fermat number. Since that time it has expanded to include k < 1200. Mark Rodenkirch (aka rogue) has been helping Ray keep the website up to date for the past few years.
Early in 2008, PrimeGrid and Proth Search teamed up to provide a software managed distributed effort to the search. Although it might appear that PrimeGrid is duplicating some of the Proth Search effort by re-doing some ranges, few ranges on Proth Search were ever double-checked. This has resulted in PrimeGrid finding primes that were missed by previous searchers. By the end of 2008, all new primes found by PrimeGrid were eligible for inclusion in Chris Caldwell's Prime Pages Top 5000. Sometime in 2009, over 90% of the tests handed out by PrimeGrid were numbers that have never been tested. For 2010, we hope to complete our reservation to 800K and extend it to 1M.
PrimeGrid intends to continue the search indefinitely for Proth primes.
What is LLR?
The Lucas-Lehmer-Riesel (LLR) test is a primality test for numbers of the form N = k*2^n − 1, with 2^n > k. Also, LLR is a program developed by Jean Penne that can run the LLR-tests. It includes the Proth test to perform +1 tests and PRP to test non base 2 numbers. See also:
(Edouard Lucas: 1842-1891, Derrick H. Lehmer: 1905-1991, Hans Riesel: 1929-2014).
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
A Little Perspective
The success of last year's Winter Solstice Challenge was incredible. It was by far the most active and productive Challenge in the two years of the series. A large part of that was because of the introduction of the CUDA application...compliments of Bryan Little (mfl0p) and Gerrit Slomma (roadrunner_gs). Also, the improvements to the AP26 CPU application allowed for credit to be at a premium (although we lowered it marginally).
It also helps having the top 2 teams tied going into the final Challenge. :)
The Challenge saw many new users and more activity in teams already participating. To those, this next Challenge may come as a shock. There are no GPU applications available. Additionally, there is no advantage of 64 bit over 32 bit. LLR is strictly a 32 bit CPU application (although it's sent to 64 bit machines). It has not benefited from the numerous optimizations that the sieves and AP26 applications have. Primality proving is a much different beast and optimization is hard to come by. However, with that said, if there are any adventurous users who wish to investigate optimizing LLR, the source can found here.
Additionally, there is a known LLR slowdown across multiple cores. This has stumped a lot of people. At this time, no one has been able to identify the exact reason. If you wish to participate in helping to identify the cause please see this thread: LLR slowdown across multiple cores.
This year's Challenge Series is a sampling of 9 different projects with a nice mixture of LLR and AP26/sieve. We consider this year's series the most challenging yet because users and teams will need to come up with different strategies to optimize their chances at winning...or at placing in the points (an LLR challenge vs. a sieve challenge vs. the AP26 challenge). Each one will have different strengths and weaknesses.
We look forward to the Challenges and to hopefully finding several major primes this year. We wish everyone the Best of Luck.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
How to Join the Challenge
To participate in the Challenge, please select only the Proth Prime Search (LLR) project in your PrimeGrid preferences section. The challenge will begin 14 February 2010 18:00 UTC and end 17 February 2010 18:00 UTC. Only work units issued AFTER 14 February 2001 18:00 UTC and received BEFORE 17 February 2001 18:00 UTC will be considered for credit.
NOTE: Only PPS work will be released during the Challenge. However, PPSE resends will occur. All completed work, PPS or PPSE, will be credited in the Challenge score.
Application builds are available for the following:
- Microsoft Windows (98 or later) running on an Intel x86-compatible CPU
- Linux running on an Intel x86-compatible CPU
- Mac OS 10.4 or later running on Intel
Note: 32bit applications sent to 64bit hosts
ATTENTION: The primality program LLR is CPU intensive; so, it is vital to have a stable system with good cooling. It does not tolerate "even the slightest of errors." Please see this post for more details on how you can "stress test" your computer. WU's will take ~5 minutes on fast/newer computers and 15+ minutes on slower/older computers. If your computer is highly overclocked, please consider "stress testing" it. :)
Time zone converter:
The World Clock - Time Zone Converter
NOTE: The countdown clock on the front page uses the host computer time. Therefore, if your computer time is off, so will the countdown clock. For precise timing, use the UTC Time in the data section to the left of the countdown clock.
Scoring Information
Scores will be kept for individuals and teams. Only work units issued AFTER 14 February 2010 18:00 UTC and received BEFORE 17 February 2010 18:00 UTC will be considered for credit. We will use the "prime score" method which is based on the n value (k*b^n-1) to score the challenge. The only difference is that the primary and double checker of a WU will receive the same score.
Therefore, each completed WU will earn a unique score based on its n value. The higher the n, the higher the score. This is different than BOINC cobblestones! A quorum of 2 is NOT needed to award Challenge score - i.e. no double checker. Therefore, each returned result will earn a Challenge score. Please note that if the result is eventually declared invalid, the score will be removed.
For details on how the score is calculated, please see this thread.
Stats will be available on the front page next to the timer within 30 minutes after the start.
What to expect at Challenge start
The server always gets hit hard at the beginning of the Challenge. However, we will be using the same procedure that we used last year for all the Challenges which proved very successful. A few hours before the start (14 February 2010 18:00 UTC) "max cache" will be dropped to 16 tasks per core. Additionally, "max to send" will be dropped to 8 tasks per core. Anyone attempting to request more will be met with "max cache" messages.
We'll raise the buffer to at least 10K. At the start of the Challenge, these settings will be adjusted as necessary. However, if the past year is any indication, we'll make the first adjustment around 30 minutes post start by raising "max cache" to 32. We'll continue to increase "max cache" when the server can handle it until a full 3 days max cache is possible.
This method allows the greatest opportunity for the most clients to get work. Even if it's just 1 task for the first 5 minutes, at least the client is crunching. Past Challenges have shown that clients have a good chance to reach "max cache" before their first task is complete. We expect the same this time.
Strategies for starting the Challenge
Depending on a variety of factors, different strategies work for different users. Here are just a few to consider:
- large farm, user can be present at start
-Set Computer is connected to the Internet about every to 0 days
-Set Maintain enough work for an additional to 0 days
-Change PrimeGrid preferences and only select PPS LLR project
-At worse, you'll only be 1 WU late on each core starting the Challenge.
-After all machines have work, increase Maintain enough work for an additional
- large farm, user NOT able to be present at start
-Same settings as above.
-At worse, you'll only be 1 WU late on each core starting the Challenge.
- a few computers, user can be present at start
-Change PrimeGrid preferences and only select PPS LLR project
-Set computers to "No New Tasks" for PrimeGrid
-At Challenge Start, update computers to "Allow New Tasks"
- a few computers, user NOT able to be present at start
-Set Computer is connected to the Internet about every to 0 days
-Set Maintain enough work for an additional to 0 days
-Change PrimeGrid preferences and only select PPS LLR project
-At worse, you'll only be 1 WU late on each core starting the Challenge.
NOTE: This presumes you have all other BOINC projects suspended.
Best of Luck to everyone!
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
24 hours to go!!!
Cupid is flying overhead while the tiger circles below. I don't know if they are going to get along but in 24 hours, we'll find out. :D
Please once again review the "Welcome" post as well as the "How to Join" post so you can be informed and make any last minute adjustments that are necessary.
A few hours before the Challenge, we begin to monitor the Meebo chat room at the bottom of the forum page. If at any time PrimeGrid becomes unresponsive, you can still access the Meebo chat room externally here: http://www.meebo.com/room/primegrid/
Before the start, the Subproject stats on the Server Status page will be disabled as will the 24 hour stats on the bottom of the front page. Also, the Top Primes Finder page will be disabled. This is all done to free up as much resources to allow the server to send and receive work.
What to expect at Challenge start
The server always gets hit hard at the beginning of the Challenge. However, a procedure that we've been using for all the Challenges this year has been very successful so far, and we hope it holds true for the final Challenge. A few hours before the start (14 February 18:00 UTC) "max cache" will be dropped to 16 tasks per core. Additionally, "max to send" will be dropped to 8 tasks per core. Anyone attempting to request more will be met with "max cache" messages.
We'll load at least 10K+ into the buffer. At the start of the Challenge, these settings will be adjusted as necessary. Historically, the first adjustment happens around 30 minutes post start by raising "max cache" to 32. We'll continue to increase "max cache" when the server can handle it until a full 3 days max cache is possible.
This method allows the greatest opportunity for the most clients to get work. Even if it's just 1 task for the first 10 minutes, at least the client is crunching. Past Challenges have shown that clients have a good chance to reach "max cache" before their first task is complete. We expect the same this time.
____________
|
|
|
 Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14044
ID: 53948
Credit: 482,370,180
RAC: 568,853
                               
|
We'll continue to increase "max cache" when the server can handle it until a full 3 days max cache is possible.
PPS LLR WUs are now coming with 26 hour deadlines. I don't think it will be possible to load up a 3 day cache since the BOINC client would go into EDF and stop requesting new tasks.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
All my clients are up since i am in the Tour de Primes. I have lowered my cache to 0 and am hoping everything will be fine since i am not at home from 1700 UTC to 2100 UTC. Hopefully i am not hit by a "deferred for x hours".
I do wonder what the standings of the Tour de Primes would be on February the 17 since some top-dogs would participate in the challenge that are not in the Tour de Primes...
____________
 |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Michael Goetz wrote: PPS LLR WUs are now coming with 26 hour deadlines. I don't think it will be possible to load up a 3 day cache since the BOINC client would go into EDF and stop requesting new tasks.
Sorry for the misinformation. Yes, you are correct. It will not be possible to have a full cache for the Challenge. This was a "last minute" change. We don't think it will have any adverse effect on the Challenge...as long as the servers remain up for 46 hours. ;)
roadrunner_gs wrote: I do wonder what the standings of the Tour de Primes would be on February the 17 since some top-dogs would participate in the challenge that are not in the Tour de Primes...
The Tour de Primes is open to anyone who completes an LLR WU during the month of February. Yes, the 17th will be very interesting. :)
____________
|
|
|
|
|
|
Hello,
has anybody the same problem?
w7, 64bit, i920
14.02.2010 08:48:32 PrimeGrid Finished download of pps_llr_44578358
14.02.2010 08:48:32 PrimeGrid Computation for task pps_llr_44578357_0 finished
14.02.2010 08:48:32 PrimeGrid Output file pps_llr_44578357_0_0 for task pps_llr_44578357_0 absent
14.02.2010 08:48:32 PrimeGrid Starting pps_llr_44578358_0
14.02.2010 08:48:32 PrimeGrid Starting task pps_llr_44578358_0 using llrPPS version 511
14.02.2010 08:48:34 PrimeGrid Computation for task pps_llr_44578358_0 finished
14.02.2010 08:48:34 PrimeGrid Output file pps_llr_44578358_0_0 for task pps_llr_44578358_0 absent
What is wrong?
____________
|
|
|
|
|
|
Im not able to get any more tasks. My client reports: "Message from server: Server can't open database". What do I do?
____________
 |
|
|
|
|
|
I have the same configuration: w7 (ultimate), 64bit, i920@3GHz, Boinc 6.10.21, running 7 tasks (plus 2 collatz).
but no problems. Seems this is a specific error on your machine. How do you run BOINC? As admin/system user, as your user, accessible for all users, protected mode, etc?
Have you any limits for your hard-disk (you can see this when you start BOINC in the message logs)
14-Feb-2010 19:59:47 [PrimeGrid] Computation for task pps_llr_44542337_1 finished
14-Feb-2010 19:59:47 [PrimeGrid] Starting pps_llr_44542311_1
14-Feb-2010 19:59:47 [PrimeGrid] Starting task pps_llr_44542311_1 using llrPPS version 511
14-Feb-2010 19:59:47 [PrimeGrid] Sending scheduler request: To fetch work.
14-Feb-2010 19:59:47 [PrimeGrid] Requesting new tasks for CPU
14-Feb-2010 19:59:49 [PrimeGrid] Started upload of pps_llr_44542337_1_0
14-Feb-2010 19:59:49 [PrimeGrid] Computation for task pps_llr_44542334_1 finished
14-Feb-2010 19:59:49 [PrimeGrid] Starting pps_llr_44542199_1
14-Feb-2010 19:59:49 [PrimeGrid] Starting task pps_llr_44542199_1 using llrPPS version 511
14-Feb-2010 19:59:50 [PrimeGrid] Finished upload of pps_llr_44542337_1_0
14-Feb-2010 19:59:51 [PrimeGrid] Started upload of pps_llr_44542334_1_0
14-Feb-2010 19:59:51 [PrimeGrid] Computation for task pps_llr_44542320_0 finished
14-Feb-2010 19:59:51 [PrimeGrid] Starting pps_llr_44542052_1
14-Feb-2010 19:59:51 [PrimeGrid] Starting task pps_llr_44542052_1 using llrPPS version 511
Hello,
has anybody the same problem?
w7, 64bit, i920
14.02.2010 08:48:32 PrimeGrid Finished download of pps_llr_44578358
14.02.2010 08:48:32 PrimeGrid Computation for task pps_llr_44578357_0 finished
14.02.2010 08:48:32 PrimeGrid Output file pps_llr_44578357_0_0 for task pps_llr_44578357_0 absent
14.02.2010 08:48:32 PrimeGrid Starting pps_llr_44578358_0
14.02.2010 08:48:32 PrimeGrid Starting task pps_llr_44578358_0 using llrPPS version 511
14.02.2010 08:48:34 PrimeGrid Computation for task pps_llr_44578358_0 finished
14.02.2010 08:48:34 PrimeGrid Output file pps_llr_44578358_0_0 for task pps_llr_44578358_0 absent
What is wrong?
____________
DeleteNull |
|
|
|
|
|
Server is now reporting down for maintenance .... we broke it.
edit: Seems to be back working now |
|
|
|
|
|
Not bad, only took 90 minutes to kill the server. :) Also, the first post says that LLR WUs will take ~5 minutes on newer CPUs. On my i7-920 running @ 2.8ghz, they are taking ~18 minutes, which is about a minute faster than my Q9450 @ 2.66ghz. What CPU is cranking out WUs in 5 minutes because I'll upgrade to it right now. :)
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
|
|
|
Hello Stoffel,
I started a virtual machine, seems to be okay. I can´t find the mistake on the main machine. Thanks.
mia7077
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14044
ID: 53948
Credit: 482,370,180
RAC: 568,853
|
Not bad, only took 90 minutes to kill the server. :) Also, the first post says that LLR WUs will take ~5 minutes on newer CPUs. On my i7-920 running @ 2.8ghz, they are taking ~18 minutes, which is about a minute faster than my Q9450 @ 2.66ghz. What CPU is cranking out WUs in 5 minutes because I'll upgrade to it right now. :)
That's way too slow for a PPS LLR. My Q6600 (Core2Quad @2.4GHz) cranks out 4 every 12 minutes. The i7's are a lot faster and ~5 min seems about right for a fast i7 with HT turned off.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
I get around 16-18 minutes per WU per thread (8 threads) on my i7 920 too.
____________
|
|
|
|
|
|
I have HT on because the performance win is too much to pass up by turning it off.
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2420
ID: 1178
Credit: 20,175,556,434
RAC: 23,180,039
                                               
|
Hello Stoffel,
I started a virtual machine, seems to be okay. I can´t find the mistake on the main machine. Thanks.
mia7077
The error's that I have been able to see (with server load isues right now, only a couple) would be consistent with GPU driver problems on your ATI 58xx series--will not affect a VM as you have discovered. When the challenge is done, you should check your drivers for problems.
____________
141941*2^4299438-1 is prime!
 |
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2420
ID: 1178
Credit: 20,175,556,434
RAC: 23,180,039
|
|
The 5 min times are for PPSE units. Right now the server is only issuing longer PPS units, which will be a good bit longer to run.
____________
141941*2^4299438-1 is prime!
|
|
|
|
|
|
My QX9770@3.8ghz is doing them in about 7mins (cpu time)
Not bad, only took 90 minutes to kill the server. :) Also, the first post says that LLR WUs will take ~5 minutes on newer CPUs. On my i7-920 running @ 2.8ghz, they are taking ~18 minutes, which is about a minute faster than my Q9450 @ 2.66ghz. What CPU is cranking out WUs in 5 minutes because I'll upgrade to it right now. :)
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14044
ID: 53948
Credit: 482,370,180
RAC: 568,853
|
I have HT on because the performance win is too much to pass up by turning it off.
For LLR tasks, it may not be worth it. For reasons not entirely understood, running multiple LLRs on the same CPU causes them all to slow down. Combined with the penalty inherent in running HT, at least one person has reported that his run times with 8 LLRs on an i7 were just under double what he was seeing running just 4 LLRs. At least on that i7, he could do 8 LLRs almost as fast with HT off as with HT on.
I guess it really comes down to this: ARE you seeing a performance win with HT on when you're running all LLRs? If you're not running all LLRs you'll see a larger gain from turning HT on.
The 5 min times are for PPSE units. Right now the server is only issuing longer PPS units, which will be a good bit longer to run.
I'm seeing 12 minutes on the PPS WUs from today's challenge. That's on a 2.4GHz C2Q, so I'm also getting hit bad with the multiple-LLR slowdown as well as running the older Core2 architecture. Also, this is my main computer, not a dedicated cruncher, and I typically see a 20% difference between CPU time and real time due to all the stuff running on this machine.
There's no way an i7 should be running 50% slower than my computer is. unless you UNDERclocked it to about 1 GHz.
Edit: Oops, I forgot Geeknik's i7 is running 8 tasks. See my post below.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
My Core2 Duo, T6500 (2.1GHz) crunches PPS LLR WU in ~12 mins. I think not bad for this machine.
____________
 |
|
|
|
|
|
My i7-965 no oc, is running them in about 13 minutes with 8 cores.. (HT on)
For yahoos, I temporarily dropped boinc manager to run only 6. It ran several in 10.5 minutes..
Yes, less time, but still not enough to cover the loss of the extra 2 virtual cores.
Also, not willing to experimanet much more atm..
I can see running with 7 if running GPU, as GPU based tasks will not need to wait for CPu bandwidth as much..
For me, HT stays on, as well as 100% cpu.
____________
|
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2420
ID: 1178
Credit: 20,175,556,434
RAC: 23,180,039
|
The 5 min times are for PPSE units. Right now the server is only issuing longer PPS units, which will be a good bit longer to run.
I'm seeing 12 minutes on the PPS WUs from today's challenge. That's on a 2.4GHz C2Q, so I'm also getting hit bad with the multiple-LLR slowdown as well as running the older Core2 architecture. Also, this is my main computer, not a dedicated cruncher, and I typically see a 20% difference between CPU time and real time due to all the stuff running on this machine.
There's no way an i7 should be running 50% slower than my computer is. unless you UNDERclocked it to about 1 GHz.
I guess we must study different math...
2.4GHz C2Q vs. 2.67GHZ i7 920 = about 90% of the clock speed
12min x 4 workunits vs. 18min x 8 workunits = 75% of the throughput
...so the i7 clearly has better performance than the C2Q even with HT turned on (my i7 is also my main home-office PC which runs a GPU project in addition to the normal home-office workload).
____________
141941*2^4299438-1 is prime!
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14044
ID: 53948
Credit: 482,370,180
RAC: 568,853
|
|
Me:
There's no way an i7 should be running 50% slower than my computer is. unless you UNDERclocked it to about 1 GHz.
Ok, I'm a dope.
I was forgetting that the i7 was running 8 tasks while I'm running only 4. So 18 minutes for a task is at 75% of the time it takes my computer. Still, that seems pretty slow compared to my machine, which was hardly state of the art when I bought it, let alone now.
Plus, my 2.4 GHz C2Q is doing them in 12 minutes and Geeknik's 2.66 GHz C2Q is doing them in 19 minutes? Something is very, very wrong there.
____________
My lucky number is 75898524288+1 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14044
ID: 53948
Credit: 482,370,180
RAC: 568,853
|
I guess we must study different math...
2.4GHz C2Q vs. 2.67GHZ i7 920 = about 90% of the clock speed
12min x 4 workunits vs. 18min x 8 workunits = 75% of the throughput
...so the i7 clearly has better performance than the C2Q even with HT turned on (my i7 is also my main home-office PC which runs a GPU project in addition to the normal home-office workload).
Yeah, I made a mistake, but I still think there's something wrong. I'm also running GPU tasks, Vista's Aero, watching videos, etc. My computer is far from optimized for crunching. The I7 isn't just higher clock speeds and HT; it's a better architecture, memory interface, and so forth and he should be crushing me.
Plus, his Q9450 -- which is the same as my CPU except with a 10% faster clock and a bigger L2 cache than mine -- is taking 50% more time to run than my computer is. Something is very much not right.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
I am at 22-23 minutes with my Xeon W3520 (2.66 GHz w/HT).
And 23-25 minutes with my Xeon E5504 (2.00 GHz w/HT).
strange.
The Q9550 (2.83 GHz) has 9.5 to 10 minutes.
The Xeon E5405 (2.0 GHz) comes in around 12.5 to 13.25 minutes. |
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2420
ID: 1178
Credit: 20,175,556,434
RAC: 23,180,039
|
Yeah, I made a mistake, but I still think there's something wrong. I'm also running GPU tasks, Vista's Aero, watching videos, etc. My computer is far from optimized for crunching. The I7 isn't just higher clock speeds and HT; it's a better architecture, memory interface, and so forth and he should be crushing me.
Yeah, I am a bit surprised, too. Mine is at 18min and is just a stock Dell XPS system, but I was hoping for a bit more with the QPI and better HT implementation (at least compared to the old HT from the P4's). Still, with the really small size of the PPS work units, the extra kick from the better bus, etc. may not be relevant like it appears to be on the bigger LLR's.
Plus, his Q9450 -- which is the same as my CPU except with a 10% faster clock and a bigger L2 cache than mine -- is taking 50% more time to run than my computer is. Something is very much not right.
Agreed...My first suspicion would be that EST or some other power saving option is accidentally turned on on his box...or maybe thermal throttling with overheating issues???
____________
141941*2^4299438-1 is prime!
|
|
|
|
|
|
So since this challenge is using prime scoring, is there a place to see how things are going scoring wise? I assume the team/individual links are tracking cobblestones. |
|
|
|
|
|
Arg, the fu**ing flash-plugin in firefox was eating up one Thread on the W3520, had to pkill firefox since it was not responding anymore. Should be faster now there.
But something is still strange:
# cat /proc/acpi/processor/CPU[0-7]/power|grep "*"
*C2: type[C3] promotion[C3] demotion[C1] latency[064] usage[01085130] duration[00000000001540708170]
*C1: type[C1] promotion[C2] demotion[--] latency[000] usage[03307072] duration[00000000000000000000]
*C1: type[C1] promotion[C2] demotion[--] latency[000] usage[02913300] duration[00000000000000000000]
*C3: type[C3] promotion[--] demotion[C2] latency[096] usage[17262851] duration[00000000051660264596]
*C2: type[C3] promotion[C3] demotion[C1] latency[064] usage[02073996] duration[00000000004362722082]
*C3: type[C3] promotion[--] demotion[C2] latency[096] usage[14009266] duration[00000000041554869523]
*C1: type[C1] promotion[C2] demotion[--] latency[000] usage[02885451] duration[00000000000000000000]
*C2: type[C3] promotion[C3] demotion[C1] latency[064] usage[01255016] duration[00000000001737446124]
on the E5504:
cat /proc/acpi/processor/CPU[0-7]/power|grep "*"
*C1: type[C1] promotion[--] demotion[--] latency[000] usage[00000000] duration[00000000000000000000]
*C1: type[C1] promotion[--] demotion[--] latency[000] usage[00000000] duration[00000000000000000000]
*C1: type[C1] promotion[--] demotion[--] latency[000] usage[00000000] duration[00000000000000000000]
*C1: type[C1] promotion[--] demotion[--] latency[000] usage[00000000] duration[00000000000000000000]
*C1: type[C1] promotion[--] demotion[--] latency[000] usage[00000000] duration[00000000000000000000]
*C1: type[C1] promotion[--] demotion[--] latency[000] usage[00000000] duration[00000000000000000000]
*C1: type[C1] promotion[--] demotion[--] latency[000] usage[00000000] duration[00000000000000000000]
*C1: type[C1] promotion[--] demotion[--] latency[000] usage[00000000] duration[00000000000000000000]
EDIT: Okay, the E5504 is an engineering sample and only supports C1 as it seems, but why are there several states in use on the W3520? |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14044
ID: 53948
Credit: 482,370,180
RAC: 568,853
|
So since this challenge is using prime scoring, is there a place to see how things are going scoring wise? I assume the team/individual links are tracking cobblestones.
Hmmm, I assumed those were the scores, not the cobblestones.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Okay the W3520 is down to constant 17.5 minutes/WU now. That sounds better.
And the standings seem to be based on the prime-score: 10.57 for a ppsLLR sounds familiar, the shorter ppseLLR are around 3.9 if i remember correctly from my last found prime this week. (or is it last week? In Germany the week starts on mondays not on sundays) |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
So since this challenge is using prime scoring, is there a place to see how things are going scoring wise? I assume the team/individual links are tracking cobblestones.
Stats are prime scores, not cobblestones. :)
____________
|
|
|
|
|
|
I calculated, by a specific workunit taking x amount of time, 8 i7 cores still beat 6 i7 cores, on WUs per hour, even though they take less time..
Another interesting thought..
In the simap races, there are several i7-920 that still beat this 965.
looking at the boink benchmarks, the floating point is higher than mine, though my integers is higher..
Then you look at the E5k series..
slower per core, but with 16 (range 12 to 32) cores... wow...
____________
|
|
|
|
|
On my i7-920 running @ 2.8ghz, they are taking ~18 minutes, which is about a minute faster than my Q9450 @ 2.66ghz. What CPU is cranking out WUs in 5 minutes because I'll upgrade to it right now. :)
Nearly all my comps are 64 bit vista or xp (the two remaining single cores are 32 bit) and the AMD 965 Deneb is doing them in 8 min; the AMD Phenom 9950's (two of them) are doing them in 10 minutes and change. only a minimal OC is in effect as I don't have time to babysit them.
____________
 |
|
|
|
|
|
Times are varying wildly here. I have 2 PPS LLR WUs ready to report that were done in 11min45sec (2.8ghz Core i7-920). And a Core2Duo powering my Windows Home Server box and it's doing PPS LLR WUs in about 14 minutes. *shrug* I don't really care. I have 3 Quad Core and 5 Dual Core doing PPS. I just want to find prime numbers. Being #1 doesn't mean anything if you're not finding prime numbers. ;)
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
|
|
So since this challenge is using prime scoring, is there a place to see how things are going scoring wise? I assume the team/individual links are tracking cobblestones.
Stats are prime scores, not cobblestones. :)
cool thanks! |
|
|
|
|
Times are varying wildly here. I have 2 PPS LLR WUs ready to report that were done in 11min45sec (2.8ghz Core i7-920). And a Core2Duo powering my Windows Home Server box and it's doing PPS LLR WUs in about 14 minutes. *shrug* I don't really care. I have 3 Quad Core and 5 Dual Core doing PPS. I just want to find prime numbers. Being #1 doesn't mean anything if you're not finding prime numbers. ;)
You certainly have the badges to show that you are looking ;)
____________
 
  |
|
|
|
|
|
What's wrong with my host
____________
 |
|
|
|
|
|
Damn, my isp has problems: no internet-connection until 2000 UTC. |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 862
ID: 18447
Credit: 881,871,632
RAC: 1,392,695
                     
|
What's wrong with my host
I don't know what's wrong as there is no stderr output shown, but it's certainly got a problem running these.
Do you have anti-virus that's not happy with the executables?
Do you have a full or write-protected disk?
I can't think of anything else to check off the top of my head.
____________
 |
|
|
|
|
|
There is this thread that is related to your problem. Maybe you should try those tips. |
|
|
|
|
|
How many Challenge score points do I get per reported PPS LLR WU?
What happens with the accumulated score after the end of the Challenge?
____________
|
|
|
|
|
|
I have all the libraries listed in that thread.
____________
|
|
|
gruenyVolunteer tester
Send message
Joined: 12 Mar 08
Posts: 22
ID: 20129
Credit: 596,251
RAC: 0
 
|
How many Challenge score points do I get per reported PPS LLR WU?
i think approx 10.
What happens with the accumulated score after the end of the Challenge?
http://www.primegrid.com/challenge/challenge.php |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
How many Challenge score points do I get per reported PPS LLR WU?
What happens with the accumulated score after the end of the Challenge?
We are testing the form k*2^n+1. Challenge score points per WU are based on the "prime score" for the given "n" tested:
Given: k*b^n+/-1
log_n = ln(k) + n*ln(b)
score = (log_n)^3 * ln(log_n) / 100,000 digit score
To find more info about "prime score", see this post.
The accumulated score will be used to determine rankings. As grueny points out, those rankings will earn Challenge Points for the overall Challenge Series.
____________
|
|
|
|
|
|
15/02/2010 21.22.57 PrimeGrid Message from server: Didn't resend lost result pps_llr_45087581_1 (expired)
15/02/2010 20.57.48 PrimeGrid Message from server: Didn't resend lost result pps_llr_45082743_1 (expired)
what's the problem?
____________
|
|
|
|
|
|
you had to complete those two WU's in three minutes and didn't, they timed out and were re-issued.. not a problem with you.
Good luck, Jack
____________
http://stats.free-dc.org/cpidtagb.php?cpid=f8f5fddb02aba8e69218c01a64840e3a&theme=8&cols=1 |
|
|
|
|
|
I assume that I will get credit from the 5 machines I have doing PPS work on PRPNet? And if so, am I getting credit now or will that be figured up after the challenge ends? Curious because my teammate has fewer cores working than I do, but he's ahead of me in the standings. ;)
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
I assume that I will get credit from the 5 machines I have doing PPS work on PRPNet? And if so, am I getting credit now or will that be figured up after the challenge ends? Curious because my teammate has fewer cores working than I do, but he's ahead of me in the standings. ;)
Sorry, PRPNet is not included in the Challenge.
From the Challenge thread:
To participate in the Challenge, please select only the Proth Prime Search (LLR) project in your PrimeGrid preferences section. The challenge will begin 14 February 2010 18:00 UTC and end 17 February 2010 18:00 UTC. Application builds are available for Linux 32 bit, Windows 32 bit and MacIntel. These applications will be sent to 64 bit clients. As with all LLR application projects, there is no advantage of 64 bit over 32 bit.
However, you will receive your "normal" PSA credit for the work. Also, any primes found will be included in the Tour de Primes.
From the Tour de Primes thread:
Prime counts and scores will be kept for all Top 5000 primes discovered in the month of February.
To participate in BOINC PPS/PPSE LLR, all you have to do is select it in your PrimeGrid preferences.
To participate in PRPNet PPSE, please see this post. Manual cobblestones are earned. However, it may take up to a month before they are applied.
____________
|
|
|
|
|
Sorry, PRPNet is not included in the Challenge.
You should probably remove the PRPNet stuff from the 1st post in this thread then. Kind of misleading to have it there and not include it in the challenge. :)
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
|
|
|
no stats-update during the last 2 hours and it's really getting tough to reach the server.. :( |
|
|
|
|
|
Some of my slower Laptops are getting more WU's then they can do before the deadline. Is there a way to adjust this?
This morning 1 of them d/l 15 WU's that were due at 7:23AM and they were d/l @ 4AM and it takes about 50 mins per WU. There was no way for it to do that many WU in that short of time.
Also what is the OLDEST BOINC that you folks recommend??? |
|
|
|
|
|
I also wonder what causes the 1 hour delay in Challenge statistics! Yesterday updates were issued every 15 min.
____________
|
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1234
ID: 18646
Credit: 920,563,862
RAC: 394,130
                
|
|
you can lower "additional work buffer" on BoincManager > Advanced > Preferences > network usage tab
because of the deadline of 1 day I ask you to lower the work buffer under 1.00 (perhaps 0.5 or 0.75)
the BoincManager asks then for less workunits
____________
Sysadm@Nbg
my current lucky number: 113856050^65536 + 1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
 |
|
|
|
|
you can lower "additional work buffer" on BoincManager > Advanced > Preferences > network usage tab
because of the deadline of 1 day I ask you to lower the work buffer under 1.00 (perhaps 0.5 or 0.75)
the BoincManager asks then for less workunits
I have them set to connect every 1 day and the additional work is set to 0.
|
|
|
|
|
|
I keep getting this error... didn't notice it until just now...
Sat 13 Feb 2010 01:46:43 PM EST PrimeGrid Message from server: No work sent
Sat 13 Feb 2010 01:46:43 PM EST PrimeGrid Message from server: PPS LLR is not available for your type of computer.
Sat 13 Feb 2010 01:46:43 PM EST PrimeGrid Message from server: No work available for the applications you have selected. Please check your settings on the web site.
|
|
|
|
|
you can lower "additional work buffer" on BoincManager > Advanced > Preferences > network usage tab
because of the deadline of 1 day I ask you to lower the work buffer under 1.00 (perhaps 0.5 or 0.75)
the BoincManager asks then for less workunits
I have them set to connect every 1 day and the additional work is set to 0.
No, other way round if you have a permanent internet connection:
connect every 0.0 day
additional work 0.5 day
With your settings you would require enough work for at least one day to cover the uptime until next connection. |
|
|
gruenyVolunteer tester
Send message
Joined: 12 Mar 08
Posts: 22
ID: 20129
Credit: 596,251
RAC: 0
|
I keep getting this error... didn't notice it until just now...
Sat 13 Feb 2010 01:46:43 PM EST PrimeGrid Message from server: No work sent
Sat 13 Feb 2010 01:46:43 PM EST PrimeGrid Message from server: PPS LLR is not available for your type of computer.
Sat 13 Feb 2010 01:46:43 PM EST PrimeGrid Message from server: No work available for the applications you have selected. Please check your settings on the web site.
i'm getting this message using a 64-bit linux. to run 32-bit applications you have to install some libs.
see http://boinc.berkeley.edu/wiki/Installing_BOINC#64_Bit_Considerations for details. |
|
|
gruenyVolunteer tester
Send message
Joined: 12 Mar 08
Posts: 22
ID: 20129
Credit: 596,251
RAC: 0
|
I keep getting this error... didn't notice it until just now...
Sat 13 Feb 2010 01:46:43 PM EST PrimeGrid Message from server: No work sent
Sat 13 Feb 2010 01:46:43 PM EST PrimeGrid Message from server: PPS LLR is not available for your type of computer.
Sat 13 Feb 2010 01:46:43 PM EST PrimeGrid Message from server: No work available for the applications you have selected. Please check your settings on the web site.
i'm getting this message using a 64-bit linux. to run 32-bit applications you have to install some libs.
see http://boinc.berkeley.edu/wiki/Installing_BOINC#64_Bit_Considerations for details. |
|
|
|
|
|
2/16/2010 10:28:43 AM|PrimeGrid|Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 4 completed tasks
2/16/2010 10:28:48 AM|PrimeGrid|Scheduler request succeeded: got 0 new tasks
2/16/2010 10:28:48 AM|PrimeGrid|Message from server: Server can't open database
|
|
|
|
|
2/16/2010 10:28:43 AM|PrimeGrid|Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 4 completed tasks
2/16/2010 10:28:48 AM|PrimeGrid|Scheduler request succeeded: got 0 new tasks
2/16/2010 10:28:48 AM|PrimeGrid|Message from server: Server can't open database
I've been getting the same thing on several of my boxes...
____________
  |
|
|
|
|
2/16/2010 10:28:43 AM|PrimeGrid|Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 4 completed tasks
2/16/2010 10:28:48 AM|PrimeGrid|Scheduler request succeeded: got 0 new tasks
2/16/2010 10:28:48 AM|PrimeGrid|Message from server: Server can't open database
I've been getting the same thing on several of my boxes...
____________
|
|
|
|
|
2/16/2010 10:28:43 AM|PrimeGrid|Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 4 completed tasks
2/16/2010 10:28:48 AM|PrimeGrid|Scheduler request succeeded: got 0 new tasks
2/16/2010 10:28:48 AM|PrimeGrid|Message from server: Server can't open database
I've been getting the same thing on several of my boxes...
____________
|
|
|
|
|
Completed result pps_llr_24233475 refused: result already reported as sucsess"
I have been getting the same error, "Server can't open database" for 1-2 hours. When I finally connected I received the message above. What does it mean and did I lose the credit for the jobs?
____________
May your addiction to Boinc be greater than mine!
End Transmission!
 |
|
|
|
|
|
I received about 75 of the messages "refused: result already reported as sucsess"
____________
May your addiction to Boinc be greater than mine!
End Transmission!
|
|
|
|
|
I received about 75 of the messages "refused: result already reported as sucsess"
THAT's no problem - they were reported, but your client did not get the message in the first time.
|
|
|
|
|
|
This happens if the server has already accepted the results, but the client does not wait up to the confirmation. Not further from meaning. |
|
|
|
|
|
and again no stats-update for several hours...
.. as is said during previous challenges: NO STATS, NO RACE.. :(
you either speed up that database by a magnitude or stop running challenges with such short WU's! |
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2420
ID: 1178
Credit: 20,175,556,434
RAC: 23,180,039
|
Completed result pps_llr_24233475 refused: result already reported as sucsess"
I have been getting the same error, "Server can't open database" for 1-2 hours. When I finally connected I received the message above. What does it mean and did I lose the credit for the jobs?
No credit lost...your result was reported, but the contact back to the client was interrupted resulting in the client thinking it still needed to report the work. When it was able to reconnect and tried reporting, the server responds with this message telling since the result was already reported successfully.
____________
141941*2^4299438-1 is prime!
|
|
|
|
|
and again no stats-update for several hours...
.. as is said during previous challenges: NO STATS, NO RACE.. :(
you either speed up that database by a magnitude or stop running challenges with such short WU's!
In der Tat. Keine Stats. grrrrr....... |
|
|
|
|
|
Thanks guys, what a relief. I am still in the race and hoping to place within the top 100. Maybe .. Maybe not?
____________
May your addiction to Boinc be greater than mine!
End Transmission!
|
|
|
|
|
Thanks guys, what a relief. I am still in the race and hoping to place within the top 100. Maybe .. Maybe not?
who knows? we are running blindfolded...
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
As y'all are fully aware, we are limping along with sporadic server overload. While it is unfortunate, it's not devastating. Most hosts are connecting within a few attempts. However, as a precautionary note, we encourage users to increase their cache to at least 4-5 hours so queues don't run dry. This hurts "prime finder" status, but it keeps cores running.
Stats now have to be run manually. There's life in the validator but it's having trouble catching up.
PRPNet has been shut down and will remain down for the duration of the Challenge. I know this puts a damper on the Tour de Primes effort but we're in survival mode now.
Everything was going well for the first 36 hours then Challenge stats runs started overwhelming the server. Once that happened, it started a snowball effect on other processes.
We are working to clean things up and decrease server load so more connections are freed up.
____________
|
|
|
|
|
|
Thanks for the update, John...You guys do what you have to to keep the server on an even keel...
____________
 |
|
|
|
|
|
Thanks John! |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14044
ID: 53948
Credit: 482,370,180
RAC: 568,853
|
|
BTW, these 26 hour deadline are wrecking havoc on my computers.
I *think* the problem is with 6.10.18, but that's just a guess.
Note that on all computers, I have only PrimeGrid running on the CPU, and only PPS-LLR tasks running. Nothing else.
Beginning as soon as I started receiving the 26 hour deadline tasks a few hours before the challenge started, BOINC decided it had to run these in High Priority mode. I'm guessing this is because it may think it can get then returned before 24 hours of the deadline -- a mere 2 hours after the WUs were sent. Note that at this time the machines had queues of 0/0.04, or just under one hour.
Oddly enough, even though running in high priority (EDF) mode, BOINC was still fetching more work. Never saw it do that before. Because it was still fetching work, and there were no other projects to preempt, the EDF mode had no effect on the computers.
Until yesterday.
(At the current time, the machines have 1 day cache settings.)
Yesterday, my quad started acting normal for EDF mode, meaning it stopped asking for new work. That's BAD. The current behavior is this:
Refuse to get new work as long as anything is running in high priority mode... which means it won't get new work until all four cores are empty. Then attempt to get a day's worth of work. If I hit a DB error, back off (and sit idle) for 1 hour before trying again. If I'm lucky enough to get a successful call to the server, I get about 2 hours of work. Then nothing gets downloaded again until the machine runs dry again.
So if I'm not babysitting the machine when it runs dry, it's going to sit idle for a long time. Best case scenario: 1 to 3 cores sit idle until the last core finishes; worst case scenario: the machine sits idle for hours and hours until it wins the lottery and gets some more work.
The two old single core machines are still downloading work despite being in EDF. They do about 2 WUs per hour, each. The quad does about 20 per hour. Unless it has no work, of course.
The short deadline is really hurting now.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
"As y'all are fully aware, we are limping along with sporadic server overload. While it is unfortunate, it's not devastating. Most hosts are connecting within a few attempts. However, as a precautionary note, we encourage users to increase their cache to at least 4-5 hours so queues don't run dry. This hurts "prime finder" status, but it keeps cores running."
I had to reduce the number of WU in order to complete them in time because of the short deadline. If I increase the number of WU's download then I am back with the same problem! |
|
|
|
|
|
Not to press the issue, but how often can we hope to see challenge stats updates?
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Thank you for the feedback on the short deadlines. It's obvious that's causing some unique problems as well.
The only good news about that is we're less than 26 hours from the finish so hopefully the problem becomes less the close we get to the finish. You probably won't see any real improvements until about 12 hours left...and that's only if you have at least 12 hours cache.
As for stats, we are presently working on them. We'll get them out as soon as possible.
____________
|
|
|
|
|
Thank you for the feedback on the short deadlines. It's obvious that's causing some unique problems as well..
you did not know about that?
48 hours most likely would have been fine, but all those hosts running them newer clients are on immediate report now.
not too late to change that deadline-setting...
|
|
|
|
|
|
It would have been better if you had started out with the deadline set to Feb 17 1900 hours.
That way You would not had to worry about making any deadline but the last one. Plus people like me would not had to abort WU's because too many were d/l to begin with!
As it is I have to set my PC's so they only get enough work to run about 4 hours!
|
|
|
HAmstyVolunteer tester
Send message
Joined: 26 Dec 08
Posts: 132
ID: 33421
Credit: 12,510,712
RAC: 0
           
|
|
so far no real problems on my side, just ran out of work due to my low cache and the unregular stats update bother me either. great work @Rytis,John and Lennart
____________
 |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Stats have been updated and will hopefully update every 1/2 hour now. WU deadlines have been extended to 48 hours but the current buffer must empty first to get to them. It's probably too little too late.
____________
|
|
|
|
|
|
Damn i got one prime every other day in the "Tour de Primes" but in this challenge and i have more clients now. I am a little unlucky at the challenges. |
|
|
|
|
Thanks guys, what a relief. I am still in the race and hoping to place within the top 100. Maybe .. Maybe not?
Same here...had just crept into the top 100, then lost ground due to a low cache setting (grr!). At 108 now and hoping to place.
On another note, my WUs are getting noticeably longer. We must really be cranking through n values!
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
On another note, my WUs are getting noticeably longer. We must really be cranking through n values!
Yes, we've gone from n~710K to n~767K. Simply AMAZING!!!
____________
|
|
|
|
|
Yes, we've gone from n~710K to n~767K. Simply AMAZING!!!
Whoa, that must be like a month's worth of normal work!
____________
|
|
|
|
|
|
Well Its not looking good for me to place in the top 100 this challenge. I do think lowering the challenge to 50 teams and 100 participants has helped competition. I will however fight to the end trying!!
____________
May your addiction to Boinc be greater than mine!
End Transmission!
|
|
|
|
|
Well Its not looking good for me to place in the top 100 this challenge. I do think lowering the challenge to 50 teams and 100 participants has helped competition. I will however fight to the end trying!!
Wow, There are 4 crunchers with out a team, If they formed there own team they would be in team points but not Individual points. But DaveSun (ranked 101) is just out of the Individual point (as of now) Real close to having point in both if only he formed his own team, [of one].
Crunch on,
Steve
Go .... SETI.USA
____________
From the High Desert in New Mexico
 |
|
|
|
|
|
I'm shutting down 5 hosts (well, one will be on if I'm home), so that should help lighten up the load. Still have the 4 fastest quads running though.
____________
|
|
|
|
|
|
At least your nodes are not hidden.
I hate the guys that participate with hidden hosts, you do not know what they have, it is as if they would start in a real sports competition under a cloak and would not let the competitors sneak a peek on their physical appearance. |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
12 hours remain!
Now is the time to start preparing for the end of the Challenge. We are lowering max cache as well as the buffer to help minimize abandoned WU's at the end. For those "moving on" after the Challenge AND have a full cache of work now (12 hours worth), it might be a good time to set "no new work". Make sure to update your clients early and often to get those completed WU's reported and counted.
There was a "rough patch" starting at hour 36 and ending at 48 where the server was in a frequent state of overload. However, for the past 12 hours, it's been smooth sailing again. :)
A quorum of 2 is NOT needed to award Challenge score - i.e. no double checker. Each returned result will earn a Challenge score. Please note that if the result is eventually declared invalid, the score will be removed. Therefore, the Challenge will not be considered final until all quorum's affecting scoring positions are completed. Because of this, we have a "clean up" period where the goal is to complete these quorums.
We hope for a fast "clean-up". If everyone can remember to either complete their cache or ABORT their WU's, then we'll be able to finalize earlier rather than later.
If you have the resources, please consider helping in the "clean-up" process after the end. However, we fully understand the need/desire to "move on."
Good Luck to everyone as we finish these last remaining hours...
At the Conclusion of the Challenge
We kindly ask users "moving on" to ABORT their WU's instead of DETACHING, RESETTING, or PAUSING.
ABORTING WU's allows them to be recycled immediately; thus a much faster "clean up" to the end of an LLR Challenge. DETACHING, RESETTING, and PAUSING WU's causes them to remain in limbo until they EXPIRE. Therefore, we must wait until WU's expire to send them out to be completed.
Please consider either COMPLETING what's in the queue or ABORTING them. Thank you. :)
____________
|
|
|
|
|
At least your nodes are not hidden.
I hate the guys that participate with hidden hosts, you do not know what they have, it is as if they would start in a real sports competition under a cloak and would not let the competitors sneak a peek on their physical appearance.
I look at it more like doing the cannonball run... I show up in a jalopy but I am not gonna let you look under the hood at my dual turbo V12 and oh yea I got a NOS just in case I need a boost. |
|
|
STE\/EVolunteer tester
Send message
Joined: 10 Aug 05
Posts: 573
ID: 103
Credit: 3,668,986,085
RAC: 203,765
     
|
At least your nodes are not hidden.
I hate the guys that participate with hidden hosts, you do not know what they have, it is as if they would start in a real sports competition under a cloak and would not let the competitors sneak a peek on their physical appearance.
I Hate the Guys that Hate the Guys that Hide their Computers ... :P
|
|
|
|
|
|
Hadn't thought about forming my "own team of one" - but have done so now. Too late for this challenge though! |
|
|
|
|
|
;-) I`m just in to say Hei!
..in case you have not noticed I`m here.. ;-))
Just few hourse left of the race - Then I shall
try to sleep a bit.. ;-D
____________
   |
|
|
|
|
[quote]At least your nodes are not hidden.
I hate the guys that participate with hidden hosts, you do not know what they have, it is as if they would start in a real sports competition under a cloak and would not let the competitors sneak a peek on their physical appearance.
There is only One reason I hide my computers;
That none of you guys shall get too dizzy by
laughing of them.. ;-)) I wouldnt like to be guilty
in other`s breakdown. :p
____________
|
|
|
|
|
12 hours remain!
... However, for the past 12 hours, it's been smooth sailing again. :)...
See....
All I had to do was to stop 5 hosts from getting work and the server load dropped enough to supply everyone else with work. I didn't think that dropping 5 hosts would fix it, but the evidence seems clear. It was I that broke the server.... LOL
Next, I'll derive the evidence for Global warming as I shovel the snow from my South Carolina decking.
Yes, this really was on my SC deck railing a couple days ago. |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
3 hours remain!
Final preparations are in full force now. We continue to lower max cache as well as the buffer to help minimize abandoned WU's at the end. For those "moving on" after the Challenge AND have a full cache of work now (3 hours worth), it might be a good time to set "no new work".
The incidence of server overload is increasing. Please update your clients early and often to get those completed WU's reported and counted. Don't wait till the end with a large cache of results to submit.
-----------------------------------------------------
A quorum of 2 is NOT needed to award Challenge score - i.e. no double checker. Each returned result will earn a Challenge score. Please note that if the result is eventually declared invalid, the score will be removed. Therefore, the Challenge will not be considered final until all quorum's affecting scoring positions are completed. Because of this, we have a "clean up" period where the goal is to complete these quorums.
We hope for a fast "clean-up". If everyone can remember to either complete their cache or ABORT their WU's, then we'll be able to finalize earlier rather than later. Expect up to a week or more before results become final.
If you have the resources, please consider helping in the "clean-up" process after the end. However, we fully understand the need/desire to "move on."
Good Luck to everyone as we finish these last remaining hours...
At the Conclusion of the Challenge
We kindly ask users "moving on" to ABORT their WU's instead of DETACHING, RESETTING, or PAUSING.
ABORTING WU's allows them to be recycled immediately; thus a much faster "clean up" to the end of an LLR Challenge. DETACHING, RESETTING, and PAUSING WU's causes them to remain in limbo until they EXPIRE. Therefore, we must wait until WU's expire to send them out to be completed.
Please consider either COMPLETING what's in the queue or ABORTING them. Thank you. :)
____________
|
|
|
|
|
12 hours remain!
... However, for the past 12 hours, it's been smooth sailing again. :)...
See....
All I had to do was to stop 5 hosts from getting work and the server load dropped enough to supply everyone else with work. I didn't think that dropping 5 hosts would fix it, but the evidence seems clear. It was I that broke the server.... LOL
Next, I'll derive the evidence for Global warming as I shovel the snow from my South Carolina decking.
Yes, this really was on my SC deck railing a couple days ago.
Are that inches or centimetres? |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14044
ID: 53948
Credit: 482,370,180
RAC: 568,853
|
|
Now this is bizarre. :)
I'm nowhere close to making the top 100 individuals (I'm at about 275 right now), but if I made my own "team-of-one", I would almost make the top 50 teams all by myself. I'd rank 54 on the teams list.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Inches and the end is on top of the rail, not beneath it. Also, it was after 9 am so it had already melted a bit. It's all gone now. |
|
|
|
|
|
We have over 20 in. in N.E. Ohio and it's not melting anytime soon. Snowing now too !! The low temps keep my basement cool and two of my P.C.s .
Good luck all, Jack |
|
|
|
|
|
Individuals should only form teams if they want others to join and compete as a team. A small team, say "old computers unite" should be able to compete in the teams even if together they would only place 300th in the individual scores. I have seen a number of individuals competing for the 100th individual position this year and the competition seems to be harder to place in than last year. Last year I could have placed in 80's but did not try because I could not get any challenge score in the individual. This year they lowered it to 100. Still a little too high for me but I tried. If someone wants to be a team of one maybe he will acquire some members and be a real team. Let individuals compete as individuals and teams as teams!
____________
May your addiction to Boinc be greater than mine!
End Transmission!
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 862
ID: 18447
Credit: 881,871,632
RAC: 1,392,695
|
|
You can create a one-man team if you wish - all teams are born this way :-)
For this challenge, I'm the only one from my team participating, though it isn't always that way.
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14044
ID: 53948
Credit: 482,370,180
RAC: 568,853
|
Individuals should only form teams if they want others to join and compete as a team.
I wasn't suggesting otherwise. Just pointing out that the competition to get on the board as an individual is much more intense than it is for teams. Most of the individuals (at least as measured by points) are concentrated in several large teams, and the size of the teams drops off very rapidly.
Also interesting is that the #1 individual would rank #8 on the teams list. Also interesting is that 1, 2, and 5 on the individual list comprise about 40% of the #1 team -- which has nearly double the points of the #2 team.
____________
My lucky number is 75898524288+1 |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Finished!!!
The Year of the Tiger Challenge has completed. The battle for primes is over. As soon as the dust settles, we'll be back with preliminary results.
Well done everyone. Once again, an incredible amount of work has been completed.
____________
|
|
|
|
|
|
So how far behind are the prime notifications? I have 10 hits but I don't know if I am the primary or the secondary yet and I have only had one 2 emails so far.
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
So how far behind are the prime notifications? I have 10 hits but I don't know if I am the primary or the secondary yet and I have only had one 2 emails so far.
Primes are lagging several hours behind. They all have to be tested for (x)gfn divisibility and that's the hold up. We are adding more resources to that now.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Preliminary data
While the Challenge is officially over, the "clean up" process now begins. If you have the resources, please consider helping. All that is required is to continue crunching PPS LLR to complete the aborted and timed-out resends. However, we fully understand the need/desire to "move on."
Official results are probably several days away, but here is some preliminary data to ponder. :)
Over 2M tasks (1M WU's) were completed resulting in over 23M cobblestones awarded...an astounding amount of work done for an LLR Challenge. Over 90 "Top 5000" primes were found. I'm sure there's more to come once the validator catches up.
254 teams and 1849 individuals participated in the Challenge.
There was a "rough patch" starting at hour 36 and ending at 48 where the server was in a frequent state of overload. However, the last 24 hours, it was smooth sailing with a little bit of turbulence in the last hour. :)
Thank you to everyone. We appreciate your participation and hope you had fun! What a GREAT start to the 2010 Challenge Series.
I think it is safe to predict the top 5 in each category. Note: scores are not final as tasks later determined to be invalid will be deducted from the total score.
Top 5 Teams
1 SETI.USA 4417339.33
2 SETI.Germany 2521777.07
3 GPU Force 1596178.85
4 BOINCstats 1187410.72
5 Team 2ch 1015281.71
Top 5 Individuals
1 Mr. Hankey SETI.USA 943517.29
2 Mumps SETI.USA 738973.74
3 j2satx US Navy 550729.16
4 [SG]marodeur6 SETI.Germany 465018.58
5 Kevint SETI.USA 391175.82
____________
|
|
|
|
|
Top 5 Teams
1 SETI.USA 4417339.33
2 SETI.Germany 2521777.07
3 GPU Force 1596178.85
4 BOINCstats 1187410.72
5 Team 2ch 1015281.71
and unoffically of course, 6-10
6 Sicituradastra. 969003.90
7 PBToyz 964487.21
8 Xtrem Team Boinc Addicted 861383.10
9 US Navy 859416.87
10 Ukraine 763439.19
Thanks for the challenge ... I enjoyed the way this worked! |
|
|
|
|
|
I keep on crunching for the "Tour de Primes".
That should help a bit while cleaning up. |
|
|
|
|
I keep on crunching for the "Tour de Primes".
That should help a bit while cleaning up.
Thank's We need many to clean up after this amazing race. :)
We did much more than i had expected.
Thanks to all 
Lennart
|
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2420
ID: 1178
Credit: 20,175,556,434
RAC: 23,180,039
|
|
Keeping mine full tilt to help with the clean-up, also.
Thanks for a fun challenge!
____________
141941*2^4299438-1 is prime!
|
|
|
rogueVolunteer developer
Send message
Joined: 8 Sep 07
Posts: 1259
ID: 12001
Credit: 18,565,548
RAC: 0
|
|
Will you also post the number of tests completed and how much the leading edge was advanced? |
|
|
|
|
|
Switching back to PPS to help the cleanup.
____________
May your addiction to Boinc be greater than mine!
End Transmission!
|
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 941
ID: 3110
Credit: 265,287,584
RAC: 72,104
          
|
Keeping mine full tilt to help with the clean-up, also.
Thanks for a fun challenge!
Keeping mine going too, but not full tilt; it seems that overclock caused one invalid WU. (Either that or a cosmic ray made it fail.) At least it wasn't an SoB WU or something!
But I agree with the second part. :)
____________
|
|
|
|
|
|
Nice challenge again.
For some stupid reason, one of my computer was set to run boinc only on idle, and although I didn't use the computer at all over the challenge, there was not a single WU computed. Even with this computer running, there is no chance for me to reach the top100.
(Talking about single user teams) I contributed 86% of my teams score, which is very sad as my team claims to have to most powerful computers and thousands of active members. There is actually some kind of "task force" to focus on challenges, but I could not convince them to go for PrimeGrid (they are more focusing on projects with strange names like yoyo and poem). As a mathematician, I have a biased opinion when it comes to the usefulness of computing prime numbers ;) |
|
|
|
|
Keeping mine full tilt to help with the clean-up, also.
Thanks for a fun challenge!
Keeping mine going too, but not full tilt; it seems that overclock caused one invalid WU. (Either that or a cosmic ray made it fail.) At least it wasn't an SoB WU or something!
But I agree with the second part. :)
It is not always an overclock that causes an invalid...I had some on my box stock purchased computers (Dell i7, Toshiba laptop).
http://www.primegrid.com/result.php?resultid=155061734
and
http://www.primegrid.com/result.php?resultid=153355973
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Unclaimed Primes
Please help us report all the primes found during the Challenge. Check out the unclaimed primes thread.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Clean up Progress
Currently 18,880 tasks are outstanding from the Challenge. Of those, 4374 are in scoring positions...affecting the outcome in 8 team positions and 34 individual positions.
We expect these numbers to drop substantially by tomorrow. Thanks to everyone for helping out with the clean up.
History
18 Feb: 18880 WU's outstanding; 4374 in scoring positions
____________
|
|
|
|
|
|
ppsLLR is empty... |
|
|
|
|
|
http://www.primegrid.com/forum_thread.php?id=1286&nowrap=true#14616
Is this option not working ? |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 941
ID: 3110
Credit: 265,287,584
RAC: 72,104
|
|
sumbangpc: Is that option set for all your profiles?
____________
|
|
|
|
|
sumbangpc: Is that option set for all your profiles?
Yes, but seems that they need my real name.
@John: Thanks for the PM, and already replied :D |
|
|
|
|
|
My father spoke to short232 yesterday at work and he mentioned that he updated his preferences as needed. If anything else needs to be done can someone please PM/e-mail him? Would hate to see both his primes be reported to someone else :) |
|
|
|
|
|
The count of primes found on my account page ticked up by one during the challenge, but I never received any notification of any sort. Does that means I was the double checker, or is there still a backlog of processing all the primes that were found?
Is there an ETA on getting the user lists of primes back online? It still says it's offline for the challenge.
I'm still waiting on that elusive first reportable prime. I did manage to be the initial finder on one top 5000 prime, but it turned out that Primegrid had already reported that prime a few months earlier...
-JMP |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
The count of primes found on my account page ticked up by one during the challenge, but I never received any notification of any sort. Does that means I was the double checker, or is there still a backlog of processing all the primes that were found?
See this post. Since your name is not listed, you must have been the double checker...or you found a prime that was already reported.
Is there an ETA on getting the user lists of primes back online? It still says it's offline for the challenge.
As soon as Challenge stats are final, purge will begin. Once the DB gets down to a reasonable size (1-2 days later), all links will be restored.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Clean up Progress
Currently 4393 tasks are outstanding from the Challenge. Of those, 376 are in scoring positions...affecting the outcome in 1 team position and 10 individual positions.
As you can see, excellent progress was made. We'll take another look within a couple hours after 18:00 UTC today (48 hours after end of Challenge). We expect results to be final then. Thanks to everyone for helping out with the clean up.
History
19 Feb 15:15 UTC: 4393 WU's outstanding; 376 in scoring positions
18 Feb 16:50 UTC: 18880 WU's outstanding; 4374 in scoring positions
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
We are still awaiting final results of the Challenge. As most have noticed, the server is limping along with sporadic connection issues. This is a result of the DB approaching 4M results. It is recommended to keep at least a couple hours of cache to weather the occasional down times. Final results are expected later today and then purging will begin. It will take a couple of days to purge all the Challenge results but at least the server will be improving along the way. :)
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Clean up Progress
Currently 934 tasks are outstanding from the Challenge. Of those, 69 are in scoring positions...affecting the outcome in 1 team position and 3 individual positions.
Clearly the ~90 that needed to get done, didn't. Of the 69 in scoring positions, really only 13 tasks need to complete to solidify standings.
Challenge stats remain disabled and will be for the duration of the clean up.
History
20 Feb 12:08 UTC: 934 tasks outstanding; 69 in scoring positions
19 Feb 21:39 UTC: 2634 tasks outstanding; 182 in scoring positions
19 Feb 15:15 UTC: 4393 tasks outstanding; 376 in scoring positions
18 Feb 16:50 UTC: 18880 tasks outstanding; 4374 in scoring positions
EDIT: Sorry, I overwrote the previous update. Summary follows:
We disabled stats which immediately improved the connection status to the server. They will remain off till the completion of the Challenge. Of the 182 tasks outstanding in scoring positions, only about 1/2 are needed to finalize the Challenge.
____________
|
|
|
|
|
The count of primes found on my account page ticked up by one during the challenge, but I never received any notification of any sort. Does that means I was the double checker, or is there still a backlog of processing all the primes that were found?
See this post. Since your name is not listed, you must have been the double checker...or you found a prime that was already reported.
Is there an ETA on getting the user lists of primes back online? It still says it's offline for the challenge.
As soon as Challenge stats are final, purge will begin. Once the DB gets down to a reasonable size (1-2 days later), all links will be restored.
I got an email and found my 1 and only prime on the top 5000. I didnt see it mentioned on the main page. I had assumed that It occurred during the challenge since I got the email on Feb 16th |
|
|
|
|
I got an email and found my 1 and only prime on the top 5000. I didnt see it mentioned on the main page. I had assumed that It occurred during the challenge since I got the email on Feb 16th
It probably was (past tense) on the front page (bottom left). I found my one and only a few weeks ago. I saw my name there. It kept on going down the list until it was gone. It only took a few days and my name was gone. Currently, the list has 18 names or primes found.
I theorize that the quantity of newly found primes was so great during the challenge that your name might have only been there a few minutes. It might be that they aren't updated very frequently and it was off the page before it even got posted.
It was a special event for me, but the limelight didn't last long.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
Results are final!
Cupid and the Tiger were quite busy snagging all those primes. In the 72 hours of the Challenge, 110 "new" Top 5000 primes were discovered with over half of those making it into the Top 1000. Never before has the Prime Pages seen such an onslaught of high ranking primes, especially Top 1000.
Over 2M tasks (1M WU's) were completed resulting in over 23M cobblestones awarded...an astounding amount of work done for an LLR Challenge.
While the prime feast was quite fulfilling, the actual entirety of the Challenge had its problems. The connection issues starting at hour 36 and ending at hour 48 were directly related to the Challenge stats script. These connection issues had adverse effects on hosts which were competing with less than 2-3 hour cache. The dreaded "backing off for 1 hour" message hurt those hosts which had less than an hour's cache. Reworking how the script ran helped resolve this issue. Additionally, PRPNet was shut down to open more connections for the Challenge. NOTE: PrimeGrid BOINC has priority over PrimeGrid PRPNet when limited resources are available.
Additional connection issues came after the conclusion of the Challenge; therefore, they had no impact on work done during the Challenge. These were directly related to the size of the database. Even with the reworked Challenge stats script, it was taking up too many resources. Once the Challenge stats were shut down completely, the connection issues minimized.
Finally, we made a miscalculation in task deadline. This caused "high priority" issues with many hosts and did not allow a full cache to be reached. However, this did not have any significant impact on the connection issues the server experienced. Unfortunately, it did impact the hosts by not requesting more work during the "high priority" period. With the connection issues, these hosts had a harder time keeping work in their cache.
The track record for Challenge conditions had really improved from 2008 till now. It had been over a year since we experienced these connection issues during a Challenge (the Oct 2009 Harvest Festival Challenge had DB issues after the Challenge). It seems as though the issues with the current Challenge brought back some old memories. For this, we apologize.
The old adages of "Success is short lived" and "you're only as good as your last performance" certainly apply here. We'll do our best to improve next time. Conditions that create these connection issues are short WU's, quorum of 2, and extended Challenge period. This leads to an increased database size which cripples the Challenge stats generation. Looking forward, we can foresee only two Challenges that may have similar issues: Full Moon, 25 July, PPSE LLR and Leonids, 17 November, SGS LLR. Fortunately, both of these are only 24 hours which minimizes the chances of problems occurring. We plan to have other measures in place for stats before these Challenges.
We hope these problems do not eclipse the incredible amount of work that was accomplished and the never before seen rate in which high ranking primes were being found. Nor should they overshadow the amazing team and individual performances turned in.
254 teams and 1849 individuals (1839 returning valid results) participated in the Challenge.
Congratulations to SETI.USA as the top team and Mr. Hankey as the top individual. Remember, the top 100 individual and top 50 team places earn points for this Challenge.
Thank you again to everyone who hung in there and kept a positive outlook. We appreciate your participation and hope you had fun! We look forward to seeing you at the next Challenge - The Ides Of March Challenge - 15 March - 24 hours on the PSP/SoB Sieve application!
Stats: Participants | Teams
Top 10 Teams
1 SETI.USA 4414145.91
2 SETI.Germany 2519413.72
3 GPU Force 1590052.79
4 BOINCstats 1186850.05
5 Team 2ch 1015168.02
6 Sicituradastra. 968942.60
7 PBToyz 964231.87
8 US Navy 858417.89
9 Xtrem Team Boinc Addicted 855866.26
10 Ukraine 762470.70
Top 10 Individuals
1 Mr. Hankey SETI.USA 943321.60
2 Mumps SETI.USA 738947.76
3 j2satx US Navy 549898.22
4 [SG]marodeur6 SETI.Germany 465005.57
5 Kevint SETI.USA 391087.78
6 [GPU Force] 7NBI_Robert GPU Force 360622.78
7 Texas_Connection GPU Force 344544.43
8 lennart SM5YMT PrimeSearchTeam 327258.95
9 ksysju BOINC@Poland 299587.49
10 [XTBA>TSA] IvanleFou Xtrem Team Boinc Addicted 286633.22
____________
|
|
|
|
|
|
Are there no more e-mails for found prime numbers?
mia7077
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
Are there no more e-mails for found prime numbers?
All emails have been sent. The only remaining primes from the Challenge are listed here: Unclaimed Primes
____________
|
|
|
|
|
Results are final!
.......
Finally, we made a miscalculation in task deadline. This caused "high priority" issues with many hosts and did not allow a full cache to be reached. However, this did not have any significant impact on the connection issues the server experienced. Unfortunately, it did impact the hosts by not requesting more work during the "high priority" period. With the connection issues, these hosts had a harder time keeping work in their cache.
.......
In my opinion: That's not so!
I'm not running the most up to date boinc client version, but I've nowhere read that this is changed lately.
Even running in high priority the Boinc client will request new work, if the cache is not filled up to the limit.
The only influence of running high priority is that the client will not switch to other projects.
____________
 |
|
|
|
|
|
agreed.
As was stated, this challenge had actually taken a step backwards
to the "good old days" of "challenging challenges".
but i think we have already beat this horse to death,
so may it rest in peace and let's forgot it as soon as possible.
Cheers. |
|
|
|
|
|
Hello John,
wolfemancs has found out that my number was already known. Then with it the thing would be clear.
mia7077
____________
|
|
|
warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 24,054,820
RAC: 0
    
|
|
Result of the challenge: primegrid is second in the projects by number list:
http://primes.utm.edu/bios/top20.php?type=project&by=PrimesRank
A couple of weeks ago we were 3rd.
Well done!
____________
 |
|
|
|
|
|
Not bad, my 2 man team finished #30 overall and earned 41.25 points. :)
____________
FUCK THIS PROJECT AND THEIR COMMIE BULLSHIT POLITICS |
|
|
|
|
Not bad, my 2 man team finished #30 overall and earned 41.25 points. :)
Congratulations, but I just don't see any points. Where are they? :)
PS:
What's a miracle? A few minutes and now I see the score. Great!
Thanks John & others.
It was very exciting. My finger still hurts from pressing update.
____________
 |
|
|
|
|
In my opinion: That's not so!
I'm not running the most up to date boinc client version, but I've nowhere read that this is changed lately.
Even running in high priority the Boinc client will request new work, if the cache is not filled up to the limit.
The only influence of running high priority is that the client will not switch to other projects.
the problem resulted in every client trying to report every result immediately - that of course generated a lot more traffic than needed.
|
|
|
|
|
|
@John:
Now the standings are officially assessed, I notice a change in your policy as compared with the 2009 Challenge Serie.
Your policy was to skip mini-teams acting as an individual from the individual standings.
Now you did not. Is this a change in policy or you just forgot?
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
@John:
Now the standings are officially assessed, I notice a change in your policy as compared with the 2009 Challenge Serie.
Your policy was to skip mini-teams acting as an individual from the individual standings.
Now you did not. Is this a change in policy or you just forgot?
Oversight on my part. The policy has not changed. Thanks for pointing this out.
EDIT: updated
____________
|
|
|
