| Author |
Message |
|---|
|
|
|
Welcome to the 321 Blast Off Challenge
With the thought to Get Ready For the Launch we are preparing for the final count down. As the server has luckily been prevented from going supernova we can indeed safely proceed to the 321 Blast Off Challenge.
An 8 day Challenge is being offered on PrimeGrid's 321 Prime Search (LLR) application. To participate in the Challenge, please select only the 321 Prime Search (LLR) project in your PrimeGrid preferences section. The challenge will begin 1 November 2012 18:00 UTC and end 9 November 2012 18:00 UTC. Application builds are available for Linux and MacIntel 32 bit and Windows 32 and 64 bit. Hosts running 64 bit Linux or MacIntel CPU's will receive the 32 bit application. 64 bit hosts can have a small advantage over 32 bit hosts.
NOTE: In your PrimeGrid preferences section, set "Send work from any subproject if selected projects have no work" to no to guarantee that no other work will be sent.
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 8-12 hours on fast/newer computers and 25+ hours 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. :)
Restricted airflow is one of the primary reasons for overheating. Take the time to monitor the fans and review the dust buildup. 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 1 November 2012 18:00 UTC and received BEFORE 9 November 2012 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 321 Prime Search
321 Prime Search is a continuation of Paul Underwood's 321 Search (see below) which looked for primes of the form 3*2^n-1. PrimeGrid added the +1 form and continues the search up to n=25M.
Primes known for 3*2^n+1 occur at the following n (PrimeGrid's finds in bold & linked):
1, 2, 5, 6, 8, 12, 18, 30, 36, 41, 66, 189, 201, 209, 276, 353, 408, 438, 534, 2208, 2816, 3168, 3189, 3912, 20909, 34350, 42294, 42665, 44685, 48150, 54792, 55182, 59973, 80190, 157169, 213321, 303093, 362765, 382449, 709968, 801978, 916773, 1832496, 2145353, 2291610, 2478785, 5082306, 7033641
Primes known for 3*2^n-1 occur at the following n (PrimeGrid's finds in bold & linked):
1, 2, 3, 4, 6, 7, 11, 18, 34, 38, 43, 55, 64, 76, 94, 103, 143, 206, 216, 306, 324, 391, 458, 470, 827, 1274, 3276, 4204, 5134, 7559, 12676, 14898, 18123, 18819, 25690, 26459, 41628, 51387, 71783, 80330, 85687, 88171, 97063, 123630, 155930, 164987, 234760, 414840, 584995, 702038, 727699, 992700, 1201046, 1232255, 2312734, 3136255, 4235414, 6090515
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: born 1929).
About 321 Search
321 Search began in February 2003 from a post by Paul Underwood seeking help from interested parties in a prime search attempt of the form 3*2^n-1. The initial goal was to build upon the completed work at Proth Search and extend the list of known primes to an exponent of 1 million (n=1M). That was quickly achieved so they advanced their goal to finding a mega prime for which they sieved up to n=5M.
As seen on PrimeGrid's front page, that goal was achieved on 23 Mar 2008, 7:57:28 UTC, when Dylan Bennett of Canada returned a positive result for n=4235414 (3*2^4235414-1). official announcement | decimal representation
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
Please change the link at the top of the home page to point to this thread, not "An Apple a Day".
Thanks!
____________
Warped
  |
|
|
|
|
|
Check, done! :)
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
I will be crunching for "321 Blast Off". I am fairly sure we will find a new prime before it is over.
There is an error in the OP -- not 22291610. |
|
|
|
|
I will be crunching for "321 Blast Off". I am fairly sure we will find a new prime before it is over.
There is an error in the OP -- not 22291610.
Unless I'm totally misunderstanding you there is no error. 3*2^2291610+1 is in the OP, with a correct link.
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
Hi Pyrus, Paul is referring to the typo 22291610 instead of the correct 2291610. The link is OK though!
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! |
|
|
|
|
Hi Pyrus, Paul is referring to the typo 22291610 instead of the correct 2291610. The link is OK though!
Ah good find! :) Didn't quite spot that and tbh didn't thoroughly check the list. Just checked if we'd had any new finds since last years challenge. Thanks for pointing it out and I fixed it.
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
Cool Challenge. I like putting the computer to work! ;) Would not miss it.
____________
My lucky number is still 13. |
|
|
|
|
|
I've got the cores all lined up and ready for the long haul. My silver badge has been waiting for this challenge all year!
____________
Proud member of Team Aggie the Pew
"Wir müssen wissen. Wir werden wissen."
"We must know, we shall know."
- David Hilbert, 1930 |
|
|
|
|
|
The leading edge right now is 8,743,943. I think we will definitely be able to get the leading edge up to 9M during the challenge. Hopefully the clean up for this challenge does not take half near forever like long challenges on (in the case of 321, relatively) long work units usually seem to.
____________
 |
|
|
|
|
|
For some reason, my Fedora17 x64 box won't get any work....tells me that there are no app builds for 64 bit linux.....any ideas for a workaround?
____________
 |
|
|
|
|
For some reason, my Fedora17 x64 box won't get any work....tells me that there are no app builds for 64 bit linux.....any ideas for a workaround?
Install 32bit libs
Lennart |
|
|
|
|
For some reason, my Fedora17 x64 box won't get any work....tells me that there are no app builds for 64 bit linux.....any ideas for a workaround?
Install 32bit libs
Lennart
Should 'yum install glibc.i686' do it?
____________
|
|
|
|
|
For some reason, my Fedora17 x64 box won't get any work....tells me that there are no app builds for 64 bit linux.....any ideas for a workaround?
Install 32bit libs
Lennart
Should 'yum install glibc.i686' do it?
I saw a similar problem being reported in the cow forums and the solution suggested there was:
sudo apt-get install ia32-libs
Though that obviously is for debian (derivatives). Installing ia32-libs with anything of your choice should solve it.
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
For some reason, my Fedora17 x64 box won't get any work....tells me that there are no app builds for 64 bit linux.....any ideas for a workaround?
Install 32bit libs
Lennart
Should 'yum install glibc.i686' do it?
I saw a similar problem being reported in the cow forums and the solution suggested there was:
sudo apt-get install ia32-libs
Though that obviously is for debian (derivatives). Installing ia32-libs with anything of your choice should solve it.
Thanks.....will try at lunchtime.....
____________
|
|
|
|
|
|
GOGOGOGOGO! :)
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
Is there really "no work available" for 321 or do I maybe have my settings wrong?
TIA dave |
|
|
|
|
|
I dont get any WUs either... |
|
|
|
|
Is there really "no work available" for 321 or do I maybe have my settings wrong?
TIA dave
Try to wait a bit. Same happened to me, but after a moment I got work. Boinc manager said the communication between the manager and the server is delayed. |
|
|
|
|
|
and we're off. I believe there's steak on the menu tonight. |
|
|
|
|
|
Work units will flow, be patient, keep in mind a lot of people are trying to connect to the server right now. Once the initial requests are done though, it should be fine for the rest of the challenge.
____________
|
|
|
 Sysadm@Nbg Volunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1181
ID: 18646
Credit: 476,764,606
RAC: 114,382
                    
|
|
no problems on my side
got work right after 18:00:00 UTC
____________
Sysadm@Nbg
my current lucky number: 3749*2^1555697+1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
 |
|
|
|
|
|
Running here too on Mac and Linux; no problem getting work right after the start.
--Gary |
|
|
|
|
|
And we're off :) My little farm is crunching 321 right now :)
____________
 |
|
|
|
|
and we're off. I believe there's steak on the menu tonight.
The only stake you rats are gonna get are the one we'll use to roast you!
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
and we're off. I believe there's steak on the menu tonight.
The only stake you rats are gonna get are the one we'll use to roast you!
/me is gonna get some BBQ-sauce. We're gonna need it to compensate for the taste of those grilled rats ;)
____________
|
|
|
|
|
and we're off. I believe there's steak on the menu tonight.
The only stake you rats are gonna get are the one we'll use to roast you!
I believe rib-eyes, t-bones and a few porter houses are in order. And it never hurts to have very cold adult beverages for the grownups and MILK shakes for the kids.
Cheers mate :)
____________
@AggieThePew
 |
|
|
|
|
and we're off. I believe there's steak on the menu tonight.
The only stake you rats are gonna get are the one we'll use to roast you!
I believe rib-eyes, t-bones and a few porter houses are in order. And it never hurts to have very cold adult beverages for the grownups and MILK shakes for the kids.
Cheers mate :)
Oh dear! You you... rat! How dare you suggest we shake our udders at some kids! :O :P
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
I love your signature Rick, a whole lot of loving for me.
quote
Member of Aggie The Pew aka AtP (aka cow tippers) |
|
|
|
|
Oh dear! You you... rat! How dare you suggest we shake our udders at some kids! :O :P
The funny part is your admitting to having udders :)
____________
@AggieThePew
|
|
|
|
|
I love your signature Rick, a whole lot of loving for me.
quote
Member of Aggie The Pew aka AtP (aka cow tippers)
Oh I do love a good steak :)
ps - as your name suggests switching teams is never too late.
____________
@AggieThePew
|
|
|
yank
Send message
Joined: 14 May 07
Posts: 111
ID: 8367
Credit: 11,470,662,775
RAC: 26
                    
|
|
This challenge series is becoming more popular every year and at present it seems that it is taking longer and longer to get some work units. One computer took 15 minutes and 56 seconds to get work units. Another one took 24 minutes and 18 seconds. My third computer took 43 minutes and 38 seconds.
Is it possible for the project to only give out the challenge work units to those who are taking part, hold off maybe an hours before allowing the other work units to be passed out to the members who do not take part in these challenges.
US Navy PrimeGrid Team. |
|
|
|
|
|
TheDawgz are looking forward to an udderly wonderful, gently pounded, veal steak (aka DPcalf) - with an adult beverage of course!
____________
There's someone in our head but it's not us. |
|
|
|
|
This challenge series is becoming more popular every year and at present it seems that it is taking longer and longer to get some work units. One computer took 15 minutes and 56 seconds to get work units. Another one took 24 minutes and 18 seconds. My third computer took 43 minutes and 38 seconds.
Is it possible for the project to only give out the challenge work units to those who are taking part, hold off maybe an hours before allowing the other work units to be passed out to the members who do not take part in these challenges.
US Navy PrimeGrid Team.
If you have use Nvidia GPu and use ATI gpu unchecked on your prefs page you should get work almost immediately.
____________
676754^262144+1 is prime |
|
|
yank
Send message
Joined: 14 May 07
Posts: 111
ID: 8367
Credit: 11,470,662,775
RAC: 26
|
|
Thanks for the information. Will see what happens next time around.
____________
 |
|
|
|
|
|
Above the countdown on the homepage, it says "Time Left To Start" instead of "Time Left to Finish." |
|
|
|
|
|
Had two start then fail within 3 seconds
See 418110202
and 418109185
<message>
- exit code -148 (0xffffff6c)
</message>
<stderr_txt>
BOINC LLR 6.03 wrapper: starting
Can't run app: -148
This is on a Windows XP 32 bit machine
Conan
EDIT :-- All work is failing with the same error, 9 have now died, will reset the project and see what happens.
Well that didn't work, so have detached and then reattached.
If I had an app.info.xml file due to when I last ran this project I didn't check and by detaching I have now removed it anyway, so I will see what happens. My other Windows XP machine is not having a problem.
____________
 |
|
|
|
|
|
Easiest (potential) fix I can come up with: don't use app_info.xml
Other questions could be:
Do other LLR subprojects work?
Have you properly configured your app_info.xml?
For people who have no idea what app_info.xml is: it's for the highly advanced users to run custom apps. Something which is generally discouraged as you will not get automatic updates when they become available.
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
Hi
Small matter of clarifation.
I set up my boxes to download after start time, but the sent time is when the boxes log on to the server.
On my event log the dowloads clearly shows after start time.
Can I have some answers?
I have uploaded 46 tasks but they all seem to have started early.
Cheers
Ross*
____________
|
|
|
|
|
|
8 Hours in, and Team Aggie The Pew scurries proudly to rank #1... at least for another half an hour or so :-)
A nod to Mr. Cow amberjr for being one of the first two to report a WU, in the prior hour. That's blazing fast!
--Gary
____________
"I am he as you are he as you are me and we are all together"
87*2^3496188+1 is prime! (1052460 digits)
4 is not prime! (1 digit) |
|
|
|
|
Hi
Small matter of clarifation.
I set up my boxes to download after start time, but the sent time is when the boxes log on to the server.
On my event log the dowloads clearly shows after start time.
Can I have some answers?
I have uploaded 46 tasks but they all seem to have started early.
Cheers
Ross*
The BOINC event log shows the time based on the clock on your computer. I believe the server's time at the challenge start was quite accurate, so the obvious first guess would be that the clocks on your computers are fast. How much "early" do they show being started in your reported results? It's hard for a non-admin to get much further than this since your computers are hidden.
--Gary |
|
|
|
|
|
Hi
I will unhide my Boxes.
Start time here in New Zealand was 7-00am
All my tasks were download and started after that in my event logs
however looking at tasks sent on the server the time is when I requested tasks but did not get them because of "suspended by the network time perferences."
I can send you every event log.
I have already uploaded 48 tasks, I would be very sorry to lose them.
Cheers
Ross*
2/11/2012 4:58:59 a.m. | PrimeGrid | Scheduler request completed: got 6 new tasks
2/11/2012 7:10:01 a.m. | | Resuming network activity
2/11/2012 7:10:01 a.m. | PrimeGrid | Started download of llr321_165224160
2/11/2012 7:10:01 a.m. | PrimeGrid | Started download of llr321_165224161
2/11/2012 7:10:04 a.m. | PrimeGrid | Finished download of llr321_165224160
2/11/2012 7:10:04 a.m. | PrimeGrid | Finished download of llr321_165224161
2/11/2012 7:10:04 a.m. | PrimeGrid | Started download of llr321_165224256
2/11/2012 7:10:04 a.m. | PrimeGrid | Started download of llr321_165224162
2/11/2012 7:10:04 a.m. | PrimeGrid | Starting task llr321_165224161_0 using llr321 version 101 in slot 0
2/11/2012 7:10:04 a.m. | PrimeGrid | Starting task llr321_165224160_0 using llr321 version 101 in slot 1
2/11/2012 7:10:06 a.m. | PrimeGrid | Finished download of llr321_165224256
2/11/2012 7:10:06 a.m. | PrimeGrid | Finished download of llr321_165224162
2/11/2012 7:10:06 a.m. | PrimeGrid | Started download of llr321_165224163
2/11/2012 7:10:06 a.m. | PrimeGrid | Started download of llr321_165224164
2/11/2012 7:10:06 a.m. | PrimeGrid | Starting task llr321_165224162_0 using llr321 version 101 in slot 2
2/11/2012 7:10:06 a.m. | PrimeGrid | Starting task llr321_165224256_0 using llr321 version 101 in slot 3
2/11/2012 7:10:08 a.m. | PrimeGrid | Finished download of llr321_165224163
2/11/2012 7:10:08 a.m. | PrimeGrid | Finished download of llr321_165224164
2/11/2012 7:10:08 a.m. | PrimeGrid | Starting task llr321_165224164_0 using llr321 version 101 in slot 4
2/11/2012 7:10:08 a.m. | PrimeGrid | Starting task llr321_165224163_1 using llr321 version 101 in slot 5
Ross*
____________
|
|
|
|
|
|
I've completed 2 work units but haven't seen my name in the results.
Do I need to sign up somewhere?
____________
 |
|
|
|
|
|
1. Make sure that the work units were sent to your machine after the challenge started.
2. Be patient, I think the leader board is only updated every few hours.
____________
|
|
|
|
|
I've completed 2 work units but haven't seen my name in the results.
Do I need to sign up somewhere?
No special action is required. Only units downloaded after 18:00 UTC on 1 November, and returned (valid) before 18:00 UTC on 9 November, count. As of the moment I'm typing this (8:10am UTC 2 Nov), all of the 321 units you've returned were downloaded too early. The ones you have "in progress" right now *will* count.
--Gary |
|
|
|
|
8 Hours in, and Team Aggie The Pew scurries proudly to rank #1... at least for another half an hour or so :-)
A nod to Mr. Cow amberjr for being one of the first two to report a WU, in the prior hour. That's blazing fast!
--Gary
Well that number 1 spot has been rectified for now ;)
But the rats are still going strong at place 2.
Keep it up!
____________
 |
|
|
|
|
|
Ross*,
From looking at your returned results and the log segment you posted, it looks like your computer clocks are correct and there has been no time zone miscalculation. I'm intrigued by the 2+ hour gap in network activity at the top of the log you posted, and that you "got new tasks" *before* the gap. Maybe the WUs were assigned to you but you didn't actually get them until later, but somehow the "clock started ticking" at the earlier point? It will take someone with a lot more knowledge than I have regarding boinc client and server function to explain this (and which I would consider a bug). Were you using the "network usage allowed" items on the client's "preferences" dialog to defer downloading until after the start? I've never heard of anyone doing that... the advice is usually to manually set "no new tasks" ahead of time and then "allow new tasks" after the start.
Hope you get a resolution... it will educate us all.
--Gary |
|
|
DoESVolunteer tester
Send message
Joined: 11 Oct 08
Posts: 784
ID: 30382
Credit: 74,878,735
RAC: 0
           
|
|
Just a guess but the log says -
"2/11/2012 4:58:59 a.m. | PrimeGrid | Scheduler request completed: got 6 new tasks" - as you say 7.00 is your start
A few years back there was something similar-- when the server allocates a task is the key time.
____________
Member of AtP
Shown here is an Australian native rat (Ratus Kickarsus) |
|
|
|
|
|
Hi Gary,
yes I set the network to download after the start as I was unable to be at my computers at start time.
It is a bug. The server noted the time I asked for Tasks but that is not the time they are sent / downloaded later.
I though [mistakenly] that this function was to do just what it appeared to do , download task after or between specific times.
the task said "download suspended - time of day"
Perhaps it is live and learn, but it appears I have lost 78 tasks, just as well it is a long challenge.
My first tasks are in now, but please fix that bug or is the delayed download function for something other than what I used it for.
Thanks for your feed back
Ross*
____________
|
|
|
|
|
Hi
I will unhide my Boxes.
Start time here in New Zealand was 7-00am
All my tasks were download and started after that in my event logs
however looking at tasks sent on the server the time is when I requested tasks but did not get them because of "suspended by the network time perferences."
I can send you every event log.
I have already uploaded 48 tasks, I would be very sorry to lose them.
Cheers
Ross*
2/11/2012 4:58:59 a.m. | PrimeGrid | Scheduler request completed: got 6 new tasks
2/11/2012 7:10:01 a.m. | | Resuming network activity
2/11/2012 7:10:01 a.m. | PrimeGrid | Started download of llr321_165224160
2/11/2012 7:10:01 a.m. | PrimeGrid | Started download of llr321_165224161
2/11/2012 7:10:04 a.m. | PrimeGrid | Finished download of llr321_165224160
2/11/2012 7:10:04 a.m. | PrimeGrid | Finished download of llr321_165224161
2/11/2012 7:10:04 a.m. | PrimeGrid | Started download of llr321_165224256
2/11/2012 7:10:04 a.m. | PrimeGrid | Started download of llr321_165224162
2/11/2012 7:10:04 a.m. | PrimeGrid | Starting task llr321_165224161_0 using llr321 version 101 in slot 0
2/11/2012 7:10:04 a.m. | PrimeGrid | Starting task llr321_165224160_0 using llr321 version 101 in slot 1
2/11/2012 7:10:06 a.m. | PrimeGrid | Finished download of llr321_165224256
2/11/2012 7:10:06 a.m. | PrimeGrid | Finished download of llr321_165224162
2/11/2012 7:10:06 a.m. | PrimeGrid | Started download of llr321_165224163
2/11/2012 7:10:06 a.m. | PrimeGrid | Started download of llr321_165224164
2/11/2012 7:10:06 a.m. | PrimeGrid | Starting task llr321_165224162_0 using llr321 version 101 in slot 2
2/11/2012 7:10:06 a.m. | PrimeGrid | Starting task llr321_165224256_0 using llr321 version 101 in slot 3
2/11/2012 7:10:08 a.m. | PrimeGrid | Finished download of llr321_165224163
2/11/2012 7:10:08 a.m. | PrimeGrid | Finished download of llr321_165224164
2/11/2012 7:10:08 a.m. | PrimeGrid | Starting task llr321_165224164_0 using llr321 version 101 in slot 4
2/11/2012 7:10:08 a.m. | PrimeGrid | Starting task llr321_165224163_1 using llr321 version 101 in slot 5
Ross*
Is it this wu on top here ?
Name llr321_165224160_0
Workunit 307581463
Created 1 Nov 2012 | 15:17:13 UTC
Sent 1 Nov 2012 | 15:58:58 UTC
Received 2 Nov 2012 | 1:38:47 UTC
Server state Over
Outcome Success
Lennart |
|
|
|
|
I've completed 2 work units but haven't seen my name in the results.
Do I need to sign up somewhere?
418019962 307245511 Sent 1 Nov 2012 | 6:03:25 UTC Recieved 2 Nov 2012 | 3:28:45 UTC Completed and validated 76,691.02 76,456.59 896.72 321 (LLR) v1.01
418008731 307534866 Sent 1 Nov 2012 | 5:17:05 UTC Recieved 2 Nov 2012 | 3:45:27 UTC Completed, waiting for validation 80,724.42 80,474.29 pending 321 (LLR) v1.01
You have downloaded those two to early.
Lennart |
|
|
|
|
8 Hours in, and Team Aggie The Pew scurries proudly to rank #1... at least for another half an hour or so :-)
A nod to Mr. Cow amberjr for being one of the first two to report a WU, in the prior hour. That's blazing fast!
--Gary
Looks like amberj was just the scout of a large herd behind him...
____________
676754^262144+1 is prime |
|
|
|
|
|
Hi
that batch of tasks were all downloaded to me as the log states
for example workunit 307581464 321-165224161
sent 15.58-58 utc 1-11-012
downloaded 2-11-2012 started 7.10-01am finished 7.10-04am
The confusion I have is the server recorded sending the task @ 15.58-58 UTC but did not download it till 7-10-01.
all those tasks downloaded to my boxes were after start time but the time recorded on the server is the time I reqested tasks but restricted downloading till after the start.
I simply prevented Boinc downloading any tasks till after the start time.
all my avx boxes compled 6 tasks each and after discovering the bug I aborted any task SENT ?? but not downloaded before the start.
all my avx boxes have now completed the second round of tasks.
Cheers
Ross*
Hope this
____________
|
|
|
|
|
Hi
that batch of tasks were all downloaded to me as the log states
for example workunit 307581464 321-165224161
sent 15.58-58 utc 1-11-012
downloaded 2-11-2012 started 7.10-01am finished 7.10-04am
The confusion I have is the server recorded sending the task @ 15.58-58 UTC but did not download it till 7-10-01.
all those tasks downloaded to my boxes were after start time but the time recorded on the server is the time I reqested tasks but restricted downloading till after the start.
I simply prevented Boinc downloading any tasks till after the start time.
all my avx boxes compled 6 tasks each and after discovering the bug I aborted any task SENT ?? but not downloaded before the start.
all my avx boxes have now completed the second round of tasks.
Cheers
Ross*
Hope this
2/11/2012 4:58:59 a.m. | PrimeGrid | Scheduler request completed: got 6 new tasks
What happend to those 6 tasks ?
they are downloaded from Boinc server 15:58:58
Challenge score are counted for wu downloaded from Nov 1 18:00 UTC to Nov 9 18:00 UTC and it uses the Boinc servertimes from the database.
Sent time for this wu in DB is Thu, 01 Nov 2012 15:58:58 GMT
Recieved time is Fri, 02 Nov 2012 01:38:47 GMT
The problem must be a client issue.
Lennart
|
|
|
|
|
|
I'm not seeing a link to the standings page. Could someone point me in the right direction?
Good luck to all!
yo |
|
|
|
|
|
User standings
Team standings
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
Thanks! |
|
|
|
|
|
just wondering, as I notice my PCs still haven't reported 12 completed tasks, is there a way to get boinc to report tasks immediately other than using cc_config? |
|
|
|
|
just wondering, as I notice my PCs still haven't reported 12 completed tasks, is there a way to get boinc to report tasks immediately other than using cc_config?
Setting it not to receive new work... tasks will be reported asap.
____________
676754^262144+1 is prime |
|
|
|
|
Setting it not to receive new work... tasks will be reported asap.
that has an unfortunate side-effect though. |
|
|
|
|
Setting it not to receive new work... tasks will be reported asap.
that has an unfortunate side-effect though.
does your using the cc_config.xml file have bad side effects or are you just wondering if there's another way to do the same thing. Not sure this would help any but you can also change your preferences in the network tab to have it communication at a very quick interval. of course none of this may help.. |
|
|
|
|
|
I was just looking for something that didn't require a restart of boinc mid challenge. The network is set to connect every 0 days with a work buffer of .1 days but while it fetches the tasks with .1 days to go it doesn't report the completed ones until it feels bored. |
|
|
|
|
|
yes there is a trick though it hurts me to tell you seeing how you're a rat and I'm a cow :P
You can make a file called cc_config.xml with the following content in your BOINC data directory:
<cc_config>
<options>
<report_results_immediately>1</report_results_immediately>
</options>
</cc_config>
If you want to know more check the boinc wiki
No need to restart boinc either, I think you can just force it to reread to config file from the GUI.
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2119
ID: 1178
Credit: 8,350,204,569
RAC: 5,885,651
                                    
|
No need to restart boinc either, I think you can just force it to reread to config file from the GUI.
I can verify that you do not need to restart to read the config file. Keep in mind that the placement of the menu item to do so varies by BOINC version.
|
|
|
|
|
|
thanks all, that's working nicely. |
|
|
|
|
|
Hello everybody!
I have a big problem.
I have 5 similar boxes based on AMD X6 1055T and 1090T.
They all have different runtimes.
From 19 hours to 35 hours. What is the problem?
Different task length or maybe problem with my hardware?
I remember the TRP challenge when we all had tasks with different length.
Is it the same situation here in 321 LLR?
____________
  |
|
|
|
|
|
I don't think there would be that big of a difference between the leading edge and the trailing edge unless we just passed over the leading edge with a huge FFT jump.
____________
|
|
|
|
|
Hello everybody!
I have 5 similar boxes based on AMD X6 1055T and 1090T.
They all have different runtimes.
From 19 hours to 35 hours. What is the problem?
is it 19-35 hours on each box or 19 on one and 35 on another similar one?
If the former then it's likely to be system bottlenecks causing tasks to run at different speeds. If the latter then I have no idea. |
|
|
|
|
is it 19-35 hours on each box or 19 on one and 35 on another similar one?
If the former then it's likely to be system bottlenecks causing tasks to run at different speeds. If the latter then I have no idea.
My answer will be
19 on one and 35 on another similar one
On each box I have +/- 2 hours of computing times but huge difference between very similar boxes.
____________
|
|
|
|
|
is it 19-35 hours on each box or 19 on one and 35 on another similar one?
If the former then it's likely to be system bottlenecks causing tasks to run at different speeds. If the latter then I have no idea.
My answer will be
19 on one and 35 on another similar one
On each box I have +/- 2 hours of computing times but huge difference between very similar boxes.
TheDawgz have a similar AMD 6-core system - with a similar +/- 3hrs. The difference being some w/u end up using an FFT of 480k and some use an FFT of 512k.
If TheDawgz are wrong - someone will, hopefully, correct us!
____________
There's someone in our head but it's not us. |
|
|
|
|
is it 19-35 hours on each box or 19 on one and 35 on another similar one?
If the former then it's likely to be system bottlenecks causing tasks to run at different speeds. If the latter then I have no idea.
My answer will be
19 on one and 35 on another similar one
On each box I have +/- 2 hours of computing times but huge difference between very similar boxes.
TheDawgz have a similar AMD 6-core system - with a similar +/- 3hrs. The difference being some w/u end up using an FFT of 480k and some use an FFT of 512k.
If TheDawgz are wrong - someone will, hopefully, correct us!
Memory bandwidth and speed can make a large difference along with FFT size. I have 2 AMD 1100Ts and an AMD 970 all with 8Gb of memory. 1 of my 1100Ts has 1333 7-7-7-22 timings and 321 WUs take about 17-18 hrs while the other has 1333 9-9-9-27 and take 20-21 hrs. My 970 has 1600 8-8-8-24 timing and takes 15-16 hrs (4 cores). The larger the number being tested the larger the FFT size which means more memory being used. Once the faster on chip L2 + L3 cache is saturated it starts pushing the data to DRAM which is much much slower so memory bandwidth and timings becomes very important for larger LLR sub projects. Also the more cores being used increase the amount of memory used increasing the bottleneck between L2 + L3 and DRAM. The motherboard can also make a minor difference as well.
Hope this gives a little insight on why WU times can vary (sometimes a lot) even with the same CPU on different systems.
____________
Largest Primes to Date:
As Double Checker: SR5 109208*5^1816285+1 Dgts-1,269,534
As Initial Finder: SR5 243944*5^1258576-1 Dgts-879,713
 |
|
|
|
|
|
I just checked in on the user stats for the challenge.. and find it amazing how
easy it now is to get up at the top. ;-))
"2 aegelhof 4671411.94" and all that by one computer, and wu`s running for 3,24 sec. ;-p
____________
   |
|
|
|
|
I just checked in on the user stats for the challenge.. and find it amazing how
easy it now is to get up at the top. ;-))
"2 aegelhof 4671411.94" and all that by one computer, and wu`s running for 3,24 sec. ;-p
Once the WU are invalidated they will be removed from the top.
____________
|
|
|
|
|
I just checked in on the user stats for the challenge.. and find it amazing how
easy it now is to get up at the top. ;-))
"2 aegelhof 4671411.94" and all that by one computer, and wu`s running for 3,24 sec. ;-p
Once the WU are invalidated they will be removed from the top.
That is not good enough.
It can take ages for all the wu`s to go through validation...and
in the meantime he shall sit on the top of the list - and keep on
adding these wu`s..?
Not very fun for the real crunchers...
____________
|
|
|
|
|
|
I think there is a logical error in the PG delivery system.
At the start of the challenge I enabled more work as usual with the preferences set correctly and got a new Genefer GPU plus four Genefer CPU instead of 321.
I have head somewhere of a similar thing reported. If you request work for GPU and CPU then the GPU request is handled first and the work sub project is not re-read for the CPU work, so you get the same sub project irrespective of your selections setup. I am too out of date to find the problem, it looks like a wrong sub-routine call or procedure call. It is not urgent, it can wait until the challenge is over and then some. This is just a heads up report to try and help the programmers from an ex programmer.
I just aborted the unwanted work and sent another update (for CPU only) and got the right stuff.
____________
Member team AUSTRALIA
My lucky number is 9291*2^1085585+1 |
|
|
|
|
I think there is a logical error in the PG delivery system.
At the start of the challenge I enabled more work as usual with the preferences set correctly and got a new Genefer GPU plus four Genefer CPU instead of 321.
I have head somewhere of a similar thing reported. If you request work for GPU and CPU then the GPU request is handled first and the work sub project is not re-read for the CPU work, so you get the same sub project irrespective of your selections setup. I am too out of date to find the problem, it looks like a wrong sub-routine call or procedure call. It is not urgent, it can wait until the challenge is over and then some. This is just a heads up report to try and help the programmers from an ex programmer.
I just aborted the unwanted work and sent another update (for CPU only) and got the right stuff.
This is a well known bug in the BOINC server software. It has been reported to BOINC devs, but they're not doing anything to fix it to the best of my knowledge.
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
I just checked in on the user stats for the challenge.. and find it amazing how
easy it now is to get up at the top. ;-))
"2 aegelhof 4671411.94" and all that by one computer, and wu`s running for 3,24 sec. ;-p
Once the WU are invalidated they will be removed from the top.
That is not good enough.
It can take ages for all the wu`s to go through validation...and
in the meantime he shall sit on the top of the list - and keep on
adding these wu`s..?
Not very fun for the real crunchers...
Yep I agree, it's super lame someone can end up first for a long time unjustly. However in order to fix this the validation functionality needs to be modified to automatically invalidate results with a time shorter than XYZ seconds, regardless of having met their quorum. Now I have no clue how the validator actually works, but I can imagine that this can turn into a pretty mess that makes stuff worse than it is. (Which is altogether mildly annoying but will sort itself in due time)
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
I just checked in on the user stats for the challenge.. and find it amazing how
easy it now is to get up at the top. ;-))
"2 aegelhof 4671411.94" and all that by one computer, and wu`s running for 3,24 sec. ;-p
Once the WU are invalidated they will be removed from the top.
That is not good enough.
It can take ages for all the wu`s to go through validation...and
in the meantime he shall sit on the top of the list - and keep on
adding these wu`s..?
Not very fun for the real crunchers...
Yep I agree, it's super lame someone can end up first for a long time unjustly. However in order to fix this the validation functionality needs to be modified to automatically invalidate results with a time shorter than XYZ seconds, regardless of having met their quorum. Now I have no clue how the validator actually works, but I can imagine that this can turn into a pretty mess that makes stuff worse than it is. (Which is altogether mildly annoying but will sort itself in due time)
With this person not being a member of a team there is no affect on team standings.
The results that count for any challenge are the final results not the updates during the challenge, so at the end the correct person will advance to the top.
This person is a 'real' cruncher but is producing invalid results, it also affects other sub-projects that this person is doing, but he has produced one valid result yesterday, but only one.
I know it does not look good and seems unfair to have a position you are not entitled to.
It may also lead to some to give up as they can't catch the "runaway" leader but with this being a week long challenge and as valid results start to come in a lot of invalid results will be removed from the tally with the score dropping accordingly.
Unfortunately you wont see a major drop (due the rate this person is producing invalid results) till near the end of the challenge or during the clean up, which is a pity.
One thing I can't get over is how fast an i5 can be, I have been paired with one and it completes a job in half the time my Phenom's do, guess compiler is tuned more to Intel than AMD (I can't admit it is a better processor now can I?).
I will just ignore the current leader of the pack and start from number 2 on the list and call them number 1 instead.
Conan
____________
|
|
|
|
|
|
The problems that the bad host (284648) cause go beyond the challenge standings being messed up temporarily. I know some say, "don't worry, they will be declared invalid" (which is true), or "the units are returned quickly, so they don't really delay any individual unit's validation", but consider these points:
- This host will generate a ton of results which are qualified for challenge credit, and so must be checked by someone else, and then likely a third cruncher. This will greatly extend the length of the clean-up.
- Individual hosts with a large number of results in the server's database caused major responsiveness problems for the server after the last challenge.
Hopefully some special arrangement can be made to exclude these results from the challenge (both standings and clean-up). I like the idea of an "automatic instant invalid" for WUs with a runtime less than some threshold. I don't know if this is feasible given the server s/w, but I hope at least that something similar is.
--Gary
p.s. Yeah, agree it's just no good that the top users get bumped down a notch, right when people are paying lots of attention to the standings...
p.p.s. There was another host running 321 just a day or two before the challenge started with the same misbehavior. Different user; glad it's gone. |
|
|
|
|
|
One question.
I'm not sure if I have to apply for this little contest ... Or do I have anything else to do ?
I set up my preference page to receive only 321LLR work.
But I can't find my name in the participant list ... ?
And yes. There are already a few packages submitted after the 1st of November.
Don't get it. Did I miss something ?
Frank
(Germany)
|
|
|
|
|
One question.
I'm not sure if I have to apply for this little contest ... Or do I have anything else to do ?
I set up my preference page to receive only 321LLR work.
But I can't find my name in the participant list ... ?
And yes. There are already a few packages submitted after the 1st of November.
Don't get it. Did I miss something ?
Frank
(Germany)
G'Day Frank,
No you don't need to apply to join in the challenge, you are already in it.
Yes you have returned results since the 1/11/2012 but unfortunately you downloaded them before the challenge started so they don't count in the challenge.
You have about 7 more results to go that were downloaded before the start of the challenge (you could abort those ones if you want).
Only work downloaded and returned between 1/11 and 9/11 count towards the challenge results. That is why you don't show up yet.
Conan
____________
|
|
|
|
|
|
nothing to see here... Conan beat me to it. |
|
|
|
|
Don't get it. Did I miss something ?
Frank
(Germany)
Yes, you missed that only tasks issued after 1 Nov 18:00 UTC are counted.
http://www.primegrid.com/results.php?hostid=290113&offset=0&show_names=0&state=0&appid=7 |
|
|
|
|
The problems that the bad host (284648) cause go beyond the challenge standings being messed up temporarily. I know some say, "don't worry, they will be declared invalid" (which is true), or "the units are returned quickly, so they don't really delay any individual unit's validation", but consider these points:
- This host will generate a ton of results which are qualified for challenge credit, and so must be checked by someone else, and then likely a third cruncher. This will greatly extend the length of the clean-up.
- Individual hosts with a large number of results in the server's database caused major responsiveness problems for the server after the last challenge.
Hopefully some special arrangement can be made to exclude these results from the challenge (both standings and clean-up). I like the idea of an "automatic instant invalid" for WUs with a runtime less than some threshold. I don't know if this is feasible given the server s/w, but I hope at least that something similar is.
--Gary
p.s. Yeah, agree it's just no good that the top users get bumped down a notch, right when people are paying lots of attention to the standings...
p.p.s. There was another host running 321 just a day or two before the challenge started with the same misbehavior. Different user; glad it's gone.
I disagree, the current process works.
____________
|
|
|
|
|
I just checked in on the user stats for the challenge.. and find it amazing how
easy it now is to get up at the top. ;-))
"2 aegelhof 4671411.94" and all that by one computer, and wu`s running for 3,24 sec. ;-p
Once the WU are invalidated they will be removed from the top.
That is not good enough.
It can take ages for all the wu`s to go through validation...and
in the meantime he shall sit on the top of the list - and keep on
adding these wu`s..?
Not very fun for the real crunchers...
Yep I agree, it's super lame someone can end up first for a long time unjustly. However in order to fix this the validation functionality needs to be modified to automatically invalidate results with a time shorter than XYZ seconds, regardless of having met their quorum. Now I have no clue how the validator actually works, but I can imagine that this can turn into a pretty mess that makes stuff worse than it is. (Which is altogether mildly annoying but will sort itself in due time)
With this person not being a member of a team there is no affect on team standings.
The results that count for any challenge are the final results not the updates during the challenge, so at the end the correct person will advance to the top.
This person is a 'real' cruncher but is producing invalid results, it also affects other sub-projects that this person is doing, but he has produced one valid result yesterday, but only one.
I know it does not look good and seems unfair to have a position you are not entitled to.
It may also lead to some to give up as they can't catch the "runaway" leader but with this being a week long challenge and as valid results start to come in a lot of invalid results will be removed from the tally with the score dropping accordingly.
Unfortunately you wont see a major drop (due the rate this person is producing invalid results) till near the end of the challenge or during the clean up, which is a pity.
One thing I can't get over is how fast an i5 can be, I have been paired with one and it completes a job in half the time my Phenom's do, guess compiler is tuned more to Intel than AMD (I can't admit it is a better processor now can I?).
I will just ignore the current leader of the pack and start from number 2 on the list and call them number 1 instead.
Conan
Hi, has anyone pm-d that dude? Maybe he does not know what's happening?
I once pm-d a dude who was producing a ton of invalids, and later when he responded he fixed the issue. The problem (he noted) was, that he try'd different os-s and did not check the wu-s. |
|
|
|
|
The problems that the bad host (284648) cause go beyond the challenge standings being messed up temporarily. I know some say, "don't worry, they will be declared invalid" (which is true), or "the units are returned quickly, so they don't really delay any individual unit's validation", but consider these points:
- This host will generate a ton of results which are qualified for challenge credit, and so must be checked by someone else, and then likely a third cruncher. This will greatly extend the length of the clean-up.
- Individual hosts with a large number of results in the server's database caused major responsiveness problems for the server after the last challenge.
Hopefully some special arrangement can be made to exclude these results from the challenge (both standings and clean-up). I like the idea of an "automatic instant invalid" for WUs with a runtime less than some threshold. I don't know if this is feasible given the server s/w, but I hope at least that something similar is.
--Gary
p.s. Yeah, agree it's just no good that the top users get bumped down a notch, right when people are paying lots of attention to the standings...
p.p.s. There was another host running 321 just a day or two before the challenge started with the same misbehavior. Different user; glad it's gone.
I disagree, the current process works.
No. At about 3 seconds per WU, this host could burn through ~20 "challenge qualified" WUs per core (and it's a quad core) in the last minute of the challenge. If this host was offline, these units most likely would not have been issued until after the deadline. But sorry, now they will have to be re-issued and completed before the challenge is "final". Extend this argument to the last 2, 5, 10, or more minutes of the challenge. Who's left to clean up this mess? Hosts like this MUST be banned (in whatever way technically possible, if any, in the BOINC world) as soon as they are recognized. FYI as of 1600 UTC Saturday this host has more challenge "credit" than the other 19 of the top 20 *combined*. And it is getting worse. How long will it take to work it all off?
I'd generally applaud an effort to contact the user, though I realize there is a fear of "big brother" intrusion, language barriers, and so on.
--Gary |
|
|
|
|
|
the bizarre thing is that it looks like only 1 of the 3 cores he's running is doing this as 2 of his tasks have been there for some time. Either way, I agree it's going to take a long time to clean up his mess - there's already several thousand tasks that will need to be checked. |
|
|
|
|
|
aegelhof is mailed and removed from 321 project now.
This will not effect the final score and all his test will faile.So another test is done at the same time and when both is done a third will go out. WHen the third is completed aegelhof test will be marked as invalid and then it will dissapear from the stats.
Lennart |
|
|
|
|
|
;-)
____________
|
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2119
ID: 1178
Credit: 8,350,204,569
RAC: 5,885,651
|
|
Looks like the same thing is going on with my teammate, Adam. I have e-mailed him to check on his machine. What I do know is that he is very overclocked, so that is the likely culprit in his case. Nevertheless, it is also true that both he and aegelhof are on Win 8. Can anyone verify that they are running with success on Win 8 so that we can definitively rule that out as a possible issue (just to make sure...I really think it is the overclocking issue)?
|
|
|
|
|
|
if it was overclocking it would last more than 3s or bluescreen on him. |
|
|
|
|
|
Looks likely to be related to Win 8. The computer of jonmarlow is doing the same too |
|
|
|
|
|
It would be very nice to have some work |
|
|
|
|
|
I too am wondering why I haven't received any work from this blast off challenge. I came to it a bit late, but its my understanding its only running a few days overall but I joined it on day 2 and I figured there would still be w/u available but I've received nothing. What gives??? |
|
|
|
|
It would be very nice to have some work
What I can see you have work.
Lennart
|
|
|
|
|
|
Sun Badger* and Barblovesroses can you please check the following things (even though they might seem silly).
- Do you have any tasks suspended for PrimeGrid? If so, you will not get new ones.
- Do you have a lot of work from other projects queued up?
- Have you selected 321 LLR in your preferences?
- Is the machine that's not getting work assigned to the venue for which you have selected 321 LLR?
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
I too am wondering why I haven't received any work from this blast off challenge. I came to it a bit late, but its my understanding its only running a few days overall but I joined it on day 2 and I figured there would still be w/u available but I've received nothing. What gives???
Do you get any error message ?
Post them or PM me with them.
Lennart |
|
|
|
|
|
no suspended tasks that I am aware of, assigned to 321 Blast off challenge, not assigned to excessive tasks, and have asked for update to prime grid and get message that there are no tasks available, even tried resetting the program and still no tasks available.
|
|
|
|
|
|
No error message Lennart, just a typical update that there are no current tasks available. |
|
|
|
|
no suspended tasks that I am aware of, assigned to 321 Blast off challenge, not assigned to excessive tasks, and have asked for update to prime grid and get message that there are no tasks available, even tried resetting the program and still no tasks available.
Have you allowed new tasks?
Do you have per chance selecter PPS LLR on CUDA or GCW Sieve?
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
and do you have allow cpu selected? |
|
|
|
|
|
Here is what I have for my preferences:
100
Use CPU
Enforced by version 6.10+
yes
Use ATI GPU
Enforced by version 6.10+
yes
Use NVIDIA GPU
Enforced by version 6.10+
yes
Is it OK for PrimeGrid and your team (if any) to email you?
Emails will be sent from services@primegrid.com; make sure your spam filter accepts this address.
yes
Should PrimeGrid show your computers on its web site?
yes
Default computer location
---
|
|
|
|
|
|
try switching off the GPUs |
|
|
|
|
|
OK, Did that, then did update to Primegrid. Its still not reporting any tasks available. |
|
|
|
|
|
You can try this, go into boinc manager and under projects highlight Primegrid and then select "reset project".
Only do this if you have no other tasks running since this will kill them all.
____________
|
|
|
|
|
|
I don't know why, but the reset is just sitting there doing nothing, not completing. Also, I had reset the Prime Grid Job an hour ago before starting to ask any questions of any of you about my problems about not getting any tasks hoping that it might solve my problems but it was not the answer. |
|
|
|
|
|
@ barblovesroses
Please note if you are using an account manager check it and confirm that you have cpu tasks enabled for Primegrid.
Nosferatu* |
|
|
|
|
|
CPU is enabled.
I finally did get the reset completed.
Still no new tasks generated though.
I feel like this is a lost cause for me.
I have run Primegrid jobs in the past and not that long ago, so I don't understand why this isn't running now. Very frustrating. Maybe I should detach and reattach. That is probably my next step here.
|
|
|
|
|
|
So I detached and reattached. I did get a task assigned to me by the system for Prime Grid, but not for the 321 Blast Off Challenge...it was for the PPS Sieve job which is another project I'm allocated to.
Must be all the tasks for 321 are allocated or something.
Anyway, I give up unless someone has any other great ideas. |
|
|
|
|
|
if it's fetching GPU tasks but not CPU tasks then either you don't have CPU correctly enabled, don't have 321 selected for your location or you have other tasks running on your cpu that's stopping it from requesting new tasks. What does your log say? |
|
|
|
|
|
Here are the messages from my manager following detachment and reattachment.
11/3/2012 1:29:54 PM PrimeGrid Started download of stat_primegrid.png
11/3/2012 1:29:54 PM PrimeGrid Started download of primegrid_slideshow_00.png
11/3/2012 1:29:55 PM PrimeGrid Finished download of stat_primegrid.png
11/3/2012 1:29:56 PM PrimeGrid Finished download of primegrid_slideshow_00.png
11/3/2012 1:31:32 PM PrimeGrid update requested by user
11/3/2012 1:31:33 PM PrimeGrid Sending scheduler request: Requested by user.
11/3/2012 1:31:33 PM PrimeGrid Not reporting or requesting tasks
11/3/2012 1:31:35 PM PrimeGrid Scheduler request completed
11/3/2012 1:32:25 PM PrimeGrid Resetting project
11/3/2012 1:32:25 PM PrimeGrid Detaching from project
11/3/2012 1:32:43 PM Fetching configuration file from http://www.primegrid.com/get_project_config.php
11/3/2012 1:33:08 PM PrimeGrid Master file download succeeded
11/3/2012 1:33:13 PM PrimeGrid Sending scheduler request: Project initialization.
11/3/2012 1:33:13 PM PrimeGrid Requesting new tasks for CPU and ATI
11/3/2012 1:33:14 PM PrimeGrid Scheduler request completed: got 1 new tasks
11/3/2012 1:33:16 PM PrimeGrid Started download of primegrid_tpsieve_1.38_windows_intelx86__ati13ati.exe
11/3/2012 1:33:16 PM PrimeGrid Started download of stat_primegrid.png
11/3/2012 1:33:17 PM PrimeGrid Finished download of stat_primegrid.png
11/3/2012 1:33:17 PM PrimeGrid Started download of primegrid_slideshow_00.png
11/3/2012 1:33:18 PM PrimeGrid Finished download of primegrid_tpsieve_1.38_windows_intelx86__ati13ati.exe
11/3/2012 1:33:20 PM PrimeGrid Finished download of primegrid_slideshow_00.png
11/3/2012 1:33:20 PM PrimeGrid Starting task pps_sr2sieve_53048299_0 using pps_sr2sieve version 138 (ati13ati) in slot 7
|
|
|
|
|
|
Do you also have a CUDA project selected for the used location, regardless of not having a Nvidia card in this host?
AFAIK you must select at least one project for CPU, ATI and CUDA for each in used location preferences and control the usage of CPU or GPU by the "Use CPU/Use ATI GPU/Use NVIDIA GPU" switches.
Edit:
This is what i have for my used location (no GPU used at PG)
Resource share
Determines the proportion of your computer's resources allocated to this project.
Example: if you participate in two BOINC projects with resource shares of 100 and 200,
the first will get 1/3 of your resources and the second will get 2/3. 100
Use CPU
Enforced by version 6.10+ yes
Use ATI GPU
Enforced by version 6.10+ no
Use NVIDIA GPU
Enforced by version 6.10+ no
Projects
321 Prime Search (LLR)
321 Blast off Challenge CPU
Cullen Prime Search (LLR) disabled
Prime Sierpinski Problem (LLR) disabled
Proth Prime Search (LLR) disabled
Seventeen or Bust (LLR) disabled
Sophie Germain Prime Search (LLR) disabled
The Riesel Problem (LLR) disabled
Woodall Prime Search (LLR) disabled
Cullen/Woodall Prime Search (Sieve) disabled
The Riesel Problem (Sieve) disabled
Proth Prime Search (Sieve) AMD (ATI)
Generalized Fermat Prime Search CUDA
Generalized Fermat Prime Search - World Record search disabled
Send work from any subproject if selected projects have no work no
There is enough 321 LLR work, as can be seen on the PG main page:
Available: 321 Prime Search (LLR) 452 |
|
|
|
|
|
does the ati task in "slot 7" mean you have 7 other tasks from another project already running? |
|
|
|
|
does the ati task in "slot 7" mean you have 7 other tasks from another project already running?
Shouldn't matter since it's actually asking for CPU work:
11/3/2012 1:33:13 PM PrimeGrid Requesting new tasks for CPU and ATI
Nonetheless, Barblovesroses, if you have other tasks running suspend those projects (for the CPUtasks only) just to make sure only PG is requesting work. |
|
|
|
|
|
I have a quadcore machine and generally there are 7 jobs running on my machine at one time. When the PPS job finished loading it immediately began running so I suspect it was in the 7th slot of the jobs that are normally running on my machine. I checked a while ago and there were actually 8 runnning at that time. It seems to depend upon the balance of CPU + GPU time needed by various jobs. |
|
|
|
|
|
]
Shouldn't matter since it's actually asking for CPU work:
11/3/2012 1:33:13 PM PrimeGrid Requesting new tasks for CPU and ATI
Nonetheless, Barblovesroses, if you have other tasks running suspend those projects (for the CPUtasks only) just to make sure only PG is requesting work.[/quote]
I'm sorry, but if I have to suspend all my other jobs just to hope to get this one job to run, I'd rather not run this job. Its not worth it to me. They are running just fine with no problems or complexities to get them to run on my computer but this has issues to get it to run and its not worth it to me. I have given it a good try today but if its this complex to get one subproject to run...well, its just not worthwhile. |
|
|
|
|
|
Did you check, what i wrote in post 59044? |
|
|
|
|
|
Yes, I tried that too and it didn't make any difference either. I don't know what else to try. |
|
|
|
|
|
try selecting "no new tasks" for your other projects before updating primegrid. |
|
|
|
|
|
3 of the 5 top users of this challenge are wasting WU's as you can see here:
http://www.primegrid.com/workunit.php?wuid=307673318 :-((
____________
DeleteNull |
|
|
|
|
3 of the 5 top users of this challenge are wasting WU's as you can see here:
http://www.primegrid.com/workunit.php?wuid=307673318 :-((
And they all have windows 8 in common.
____________
676754^262144+1 is prime |
|
|
|
|
|
I have a Win8 box too http://www.primegrid.com/results.php?hostid=299960&offset=0&show_names=0&state=0&appid=7, but the results need 37,000 seconds, not 3!
3 of the 5 top users of this challenge are wasting WU's as you can see here:
http://www.primegrid.com/workunit.php?wuid=307673318 :-((
And they all have windows 8 in common.
____________
DeleteNull |
|
|
|
|
I have a Win8 box too http://www.primegrid.com/results.php?hostid=299960&offset=0&show_names=0&state=0&appid=7, but the results need 37,000 seconds, not 3!
Then the issue must be somewhere else.
____________
676754^262144+1 is prime |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13389
ID: 53948
Credit: 228,191,239
RAC: 170,134
                 
|
|
Regarding Windows 8:
Just before Sandy hit I had noticed that there were several Windows 8 systems that were producing 100% errors.
I was able to verify that other Win 8 systems were correctly returning LLR tasks.
That's not proof that there isn't a problem with Win 8, but it's certain that at least some Win 8 systems are working correctly.
____________
My lucky number is 75898524288+1 |
|
|
|
|
I have a Win8 box too http://www.primegrid.com/results.php?hostid=299960&offset=0&show_names=0&state=0&appid=7, but the results need 37,000 seconds, not 3!
3 of the 5 top users of this challenge are wasting WU's as you can see here:
http://www.primegrid.com/workunit.php?wuid=307673318 :-((
And they all have windows 8 in common.
Your windows machine runs BOINC Client 7.0.36
aegelhof,
Adam,
jonmarlow
all 3 have Win 8 x64 like you but they all run BOINC Client 7.0.28 and they all are at the top with hundreds of failed work units (Adam has no successful ones at all).
Could this be the problem, the BOINC Client version?
____________
|
|
|
|
|
I have a Win8 box too http://www.primegrid.com/results.php?hostid=299960&offset=0&show_names=0&state=0&appid=7, but the results need 37,000 seconds, not 3!
3 of the 5 top users of this challenge are wasting WU's as you can see here:
http://www.primegrid.com/workunit.php?wuid=307673318 :-((
And they all have windows 8 in common.
Your windows machine runs BOINC Client 7.0.36
aegelhof,
Adam,
jonmarlow
all 3 have Win 8 x64 like you but they all run BOINC Client 7.0.28 and they all are at the top with hundreds of failed work units (Adam has no successful ones at all).
Could this be the problem, the BOINC Client version?
It could be. A member of our team is a boinc alpha tester and he strongly recommended running 7.0.36 on Win8 until something better comes along. I have one rig running win8 & 7.0.36 in the race and it seems to be OK although a few of the results are marked "validation inconclusive" but in those cases my wingman is one of those computers with 3.x seconds runtime and 0.0 seconds CPU time. Here's a link to my rig running win8/7.0.36 http://www.primegrid.com/results.php?hostid=298696 |
|
|
|
|
Could this be the problem, the BOINC Client version?
Versions before 7.0.31:
1. Installation issue on Windows 8.
____________
 |
|
|
|
|
and we're off. I believe there's steak on the menu tonight.
The only stake you rats are gonna get are the one we'll use to roast you!
I believe rib-eyes, t-bones and a few porter houses are in order. And it never hurts to have very cold adult beverages for the grownups and MILK shakes for the kids.
Cheers mate :)
I'm in on all the grill items! But, I'd like one of the shakes, myself! ;)
____________
My lucky number is still 13. |
|
|
|
|
|
The leading edge is almost at 9M, we will definitely hit this year's goal during the challenge.
____________
|
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2119
ID: 1178
Credit: 8,350,204,569
RAC: 5,885,651
|
Could this be the problem, the BOINC Client version?
Versions before 7.0.31:
1. Installation issue on Windows 8.
I e-mailed Adam and asked him to update BOINC and give it a try again (He is traveling and able to do some things remotely). Sadly, looks like no luck. He is still giving the same errors.
I am curious if maybe Windows 8 has a potential issue similar to one in Linux where the 64-bit versions do not have the 32-bit libraries by default? Maybe 3 seconds or so for the wrapper to start in Windows and then crash when 32-bit libraries aren't there?
|
|
|
|
|
I am curious if maybe Windows 8 has a potential issue similar to one in Linux where the 64-bit versions do not have the 32-bit libraries by default? Maybe 3 seconds or so for the wrapper to start in Windows and then crash when 32-bit libraries aren't there?
My windows 8 pro (64-bit) was a clean install and I didn't see an option for installing 32-bit libraries. I then installed the 64-bit version of BOINC 7.0.36. I do see the wrapper running (once instance per logical core) and all are labeled as 32-bit processes (task manager) along with other 32-bit apps including chrome. The boinc manager, client and primegrid_cllr.exe all appear to be 64-bit processes as expected, correct? So based on the above, it appears the 32-bit libraries are installed by default on 64-bit windows 8. |
|
|
|
|
|
HI all. Just got back from vacation on the weekend. So sorry everyone, I meant no harm. I just wanted to participate so I turned on my PC and left it running for the weekend while I was away... I hope the mistakes my computer made can lead to something being learned from it so that it doesn't happen to another. Again, I apologize. |
|
|
|
|
HI all. Just got back from vacation on the weekend. So sorry everyone, I meant no harm. I just wanted to participate so I turned on my PC and left it running for the weekend while I was away... I hope the mistakes my computer made can lead to something being learned from it so that it doesn't happen to another. Again, I apologize.
Hiya! :) Yeah you caused a bit of a kerfuffle but nothing that won't get solved during the clean up. In order to avoid continuing this mess please try upgrading to the very latest BOINC version.
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
Hey sorry about the confusion my machine and I contributed to. I was also traveling this weekend and set BOINC to work on the challenge while I was out so I wasn't able to catch the error before it blew up. I attempted to make some corrections like updating the BOINC client to a newer version via remote but that didn't seem to solve the issue.
I'd like to try and troubleshoot this error so we can find maybe find a solution. I am running Windows 8 pro 64bit, which was my initial concern when Scott Brown had let me know my units were not completing, given that Windows 8 was new and BOINC may not have been configured or updated properly for this. But looking over the thread I see mention that others have gotten Windows 8 to work for them so the error must be specific to me.
I am running an i5 3570k that was OCd to 4.5ghz. This was the next logical culprit since it could be unstable in this work environment though I have had no blue screens or other problems whatsoever in about the month that I was using this set up on Windows 7 64 bit or the past week of Windows 8 64 bit.
I noticed that at some point today my Preferences had been set to a different project than the 321 challenge. Thanks for that! I really don't like confusing the rankings for those who are actually competing for the top spots. I hope you find it easy enough to just assume that my name may as well not be there and count it as such until my work can be invalidated.
I tried to update my preferences this evening to work again on the 321 challenge after I had reset my OC so I could try to narrow this problem down. However, after resetting Prime Grid on BOINC and updating the project I was still receiving work from another project. Possibly my preferences haven't been communicated to the BOINC client yet but it has been about an hour of attempting this. So if there is anything else I can do or if a moderator/admin wants to contact me I'd be happy to try and work through this.
Thanks,
Adam |
|
|
|
|
|
My i5-2500k OC4.5G@1.25V with offset mode. I have used this setting for more than half year,no blue screen no error, every thing is OK.
After join 321 challenge, blue screen with the error number 124 appears every 1-2 hours! Finally I find my boinc setting 'use 100% CPU in 80% time' cause this problem.
In offset mode, 80% time CPU run in 4.5G@1.25v and 20% time 1.6G@0.9*v, the continually replacement of high and low voltage make the cpu unsteady.
So I change the setting 'use 100% CPU in 100% time', every thing become OK again.
Hope this post can help some OCer.
____________
|
|
|
|
|
|
We have officially moved the leading edge past the 9M mark, which was the goal for the year. Great job everyone.
____________
|
|
|
|
|
Hey sorry about the confusion my machine and I contributed to. I was also traveling this weekend and set BOINC to work on the challenge while I was out so I wasn't able to catch the error before it blew up. I attempted to make some corrections like updating the BOINC client to a newer version via remote but that didn't seem to solve the issue.
I'd like to try and troubleshoot this error so we can find maybe find a solution. I am running Windows 8 pro 64bit, which was my initial concern when Scott Brown had let me know my units were not completing, given that Windows 8 was new and BOINC may not have been configured or updated properly for this. But looking over the thread I see mention that others have gotten Windows 8 to work for them so the error must be specific to me.
I am running an i5 3570k that was OCd to 4.5ghz. This was the next logical culprit since it could be unstable in this work environment though I have had no blue screens or other problems whatsoever in about the month that I was using this set up on Windows 7 64 bit or the past week of Windows 8 64 bit.
I noticed that at some point today my Preferences had been set to a different project than the 321 challenge. Thanks for that! I really don't like confusing the rankings for those who are actually competing for the top spots. I hope you find it easy enough to just assume that my name may as well not be there and count it as such until my work can be invalidated.
I tried to update my preferences this evening to work again on the 321 challenge after I had reset my OC so I could try to narrow this problem down. However, after resetting Prime Grid on BOINC and updating the project I was still receiving work from another project. Possibly my preferences haven't been communicated to the BOINC client yet but it has been about an hour of attempting this. So if there is anything else I can do or if a moderator/admin wants to contact me I'd be happy to try and work through this.
Thanks,
Adam
No problem :)
It was me changing your prefs.
This does not seems to be a problem only on your host.
Lennart
|
|
|
|
|
My i5-2500k OC4.5G@1.25V with offset mode. I have used this setting for more than half year,no blue screen no error, every thing is OK.
After join 321 challenge, blue screen with the error number 124 appears every 1-2 hours! Finally I find my boinc setting 'use 100% CPU in 80% time' cause this problem.
In offset mode, 80% time CPU run in 4.5G@1.25v and 20% time 1.6G@0.9*v, the continually replacement of high and low voltage make the cpu unsteady.
So I change the setting 'use 100% CPU in 100% time', every thing become OK again.
Hope this post can help some OCer.
You have a quad core? Use this in- boinc\tools\computing preferences\prosessor usage, and look on the bottom: On multiprosessor systems, use at most [75,00] % of the prosessors. Default is 100%. For a quad 25% represents 1core. It will fix that problem if you want to use less cpu time for boinc or save heat. I use it myself. |
|
|
|
|
|
Two hosts with no prior invalid tasks fail to validate after task has been sent to two "rogue hosts":
http://www.primegrid.com/workunit.php?wuid=307695449
____________
676754^262144+1 is prime |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13389
ID: 53948
Credit: 228,191,239
RAC: 170,134
|
Two hosts with no prior invalid tasks fail to validate after task has been sent to two "rogue hosts":
http://www.primegrid.com/workunit.php?wuid=307695449
There's no problem there. The rogue hosts are being ignored. The two "good" hosts are still inconclusive because they returned residues that did not match each other.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
I work with Turbo Boost 8 cores, ( 2 tasks ) but thats not better
____________
|
|
|
|
|
|
I have a box with AMD X6 1090T and more than half of workunits are getting "Completed, validation inconclusive".
Some say that this workunits will not get final scores. But I don't see any computation errors while crunching these tasks.
Here's a proof:
http://www.primegrid.com/results.php?hostid=275983&offset=0&show_names=0&state=2&appid=7
And this PC also is getting validation inconclusive.
http://www.primegrid.com/results.php?hostid=299995&offset=0&show_names=0&state=2&appid=
So what to do?
____________
|
|
|
|
|
I have a box with AMD X6 1090T and more than half of workunits are getting "Completed, validation inconclusive".
Some say that this workunits will not get final scores. But I don't see any computation errors while crunching these tasks.
Here's a proof:
http://www.primegrid.com/results.php?hostid=275983&offset=0&show_names=0&state=2&appid=7
And this PC also is getting validation inconclusive.
http://www.primegrid.com/results.php?hostid=299995&offset=0&show_names=0&state=2&appid=
So what to do?
If you click on the results for the first host you'll notice that all but one are a doublechecker of the rogue Windows 8 hosts. As they didn't return a (correct) solution and yours is compared to them the only thing the server can do is conclude that either one of you is wrong and send out another unit. It's more than likely that yours will turn out valid when a third pc returns a result.
What's going on with the second host I can't speak to.
However, it is possible for a workunit to finish in such a way that BOINC thinks it's done and done properly, but the result still is invalid. See e.g. those rogue windows 8 hosts, though the same can happen with actual computations.
Also, did the hosts produce correct results on LLR projects before? LLR is fairly intensive job for a CPU and especially if it's overclocked or getting very hot computation erros (that may not be caught by the program) can occur.
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
|
The rise and fall of aegelhof, Adam and jonmarlow:
 |
|
|
|
|
|
The system works!
____________
|
|
|
|
|
The rise and fall of aegelhof, Adam and jonmarlow:
My 15 minutes of BOINC fame are ending. Fun graph to see. But I am very glad it is working itself out.
Is there any idea when the host issue may be resolved? |
|
|
|
|
My 15 minutes of BOINC fame are ending. Fun graph to see. But I am very glad it is working itself out.
Is there any idea when the host issue may be resolved?
If you install BOINC client version 7.0.36, your problem should resolve on Windows 8.
____________
|
|
|
|
|
My 15 minutes of BOINC fame are ending. Fun graph to see. But I am very glad it is working itself out.
Is there any idea when the host issue may be resolved?
If you install BOINC client version 7.0.36, your problem should resolve on Windows 8.
I'm not sure it's that simple. I was running win 8 pro 64-bit with 64-bit 7.0.36 and with last check, 40+ WUs have been validated. I decided to test WU crunch time with HT turned off (3930K processor). I rebooted into bios, turned HT off and booted into win 8. I had 12 WUs going at various stages of completion before the reboot and 7 of them completed normally with HT off. As soon as a new WU started, I saw the same thing that Adam is experiencing. 3.xx seconds WU time and 0.00 CPU time on about 7 WUs before I stopped fetching new WUs. I don't have an explanation for this but I did reboot back into linux (OpenSuse 12.2, 64-bit) and WUs have been completing and validating with or without HT. The other thing I noticed was that win 8 completed the WUs about 5-10% faster than OpenSuse. Anyone else seen this? |
|
|
|
|
|
Can I just say I'm amazed by the turnout so far? Currently 1355 people are participating the challenge. That's a whole lot of people cruncing away like at it! :)
Lets keep it going! My guts tell me a new 321 prime is around the corner!
Though I'm not entirely certain how far away that corner is
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2119
ID: 1178
Credit: 8,350,204,569
RAC: 5,885,651
|
My 15 minutes of BOINC fame are ending. Fun graph to see. But I am very glad it is working itself out.
Is there any idea when the host issue may be resolved?
If you install BOINC client version 7.0.36, your problem should resolve on Windows 8.
Adam upgraded BOINC during the run with no effect. I do not think it has to do with the BOINC version at this point.
|
|
|
|
|
|
Are the win8 guys running as administrator? If not could it be something to do with permissions on either the app or folders? |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13389
ID: 53948
Credit: 228,191,239
RAC: 170,134
|
Are the win8 guys running as administrator? If not could it be something to do with permissions on either the app or folders?
The BOINC installation options may also be important. Perhaps it works when running as a service, but not when run as an application, or vice versa.
____________
My lucky number is 75898524288+1 |
|
|
|
|
My 15 minutes of BOINC fame are ending. Fun graph to see. But I am very glad it is working itself out.
Is there any idea when the host issue may be resolved?
If you install BOINC client version 7.0.36, your problem should resolve on Windows 8.
I'm not sure it's that simple. I was running win 8 pro 64-bit with 64-bit 7.0.36 and with last check, 40+ WUs have been validated. I decided to test WU crunch time with HT turned off (3930K processor). I rebooted into bios, turned HT off and booted into win 8. I had 12 WUs going at various stages of completion before the reboot and 7 of them completed normally with HT off. As soon as a new WU started, I saw the same thing that Adam is experiencing. 3.xx seconds WU time and 0.00 CPU time on about 7 WUs before I stopped fetching new WUs. I don't have an explanation for this but I did reboot back into linux (OpenSuse 12.2, 64-bit) and WUs have been completing and validating with or without HT. The other thing I noticed was that win 8 completed the WUs about 5-10% faster than OpenSuse. Anyone else seen this?
I have a 3930K also, and I am interested in your results here. What are your results comparing HT with not HT?
____________
|
|
|
|
|
I have a 3930K also, and I am interested in your results here. What are your results comparing HT with not HT?
In linux, 28,000-30.000 seconds with HT off and 58,000-62,000 seconds with HT on.
In Win 8 with HT on, ~55,000-58,000 sec.
Not a big difference, if any, I think. |
|
|
|
|
My 15 minutes of BOINC fame are ending. Fun graph to see. But I am very glad it is working itself out.
Is there any idea when the host issue may be resolved?
If you install BOINC client version 7.0.36, your problem should resolve on Windows 8.
Adam upgraded BOINC during the run with no effect. I do not think it has to do with the BOINC version at this point.
Right, I've installed 7.0.36 and it didn't correct the issue. I've re-installed it tonight and tried running the program in Windows 7 compatibility mode, which did not help. From my troubleshooting I don't think the problem is with the BOINC client. If I'm not mistaken only those effected by this are using Windows 8 64 bit.
I did have a breakthrough as I'm writing this!
I took J.Sheridan's advice and ran BOINC as an administrator and am currently running four 321 WUs for 5 minutes without error. For disclosure I did this in Windows 7 compatibility mode originally. I have since suspended the WUs and closed BOINC. I then adjusted the compatibility mode back to not use the compatibility mode. Opening BOINC, again as administrator, the WUs were resumed successfully.
To further test this I reset the project, now with it being opened without compatibility mode as an administrator, in Windows 8 Pro 64 bit and the units are successfully chugging along with no errors after one minute.
This is all done at stock CPU settings. I'm going to go back into UEFI and reapply my OC settings and attempt this one more time to rule out any possible OC issues.
EDIT: Back up and running with no compatibility settings, BOINC as an administrator, in Windows 8 64 bit, with an i5 3570k OCd to 4.5 ghz I was able to reset the project and have been successfully running four 321 WUs for five minutes. Hopefully in 14 hours I'll be able to come back and let you all know that the units were completed without error. |
|
|
|
|
I have a 3930K also, and I am interested in your results here. What are your results comparing HT with not HT?
In linux, 28,000-30.000 seconds with HT off and 58,000-62,000 seconds with HT on.
In Win 8 with HT on, ~55,000-58,000 sec.
Not a big difference, if any, I think.
Yeah, HT is definitely not worth it. I was thinking the huge L3 cache may have helped with that, but guess not for LLR. What clock speed is yours set at? Mine is currently at 4GHz and I am just barely under 30K seconds per WU.
____________
|
|
|
|
|
Yeah, HT is definitely not worth it.
timewise, no, but if you're after credits then you get more credits/wu with HT on.
|
|
|
|
|
Yeah, HT is definitely not worth it. I was thinking the huge L3 cache may have helped with that, but guess not for LLR. What clock speed is yours set at? Mine is currently at 4GHz and I am just barely under 30K seconds per WU.
3.8 GHz w/ all 4 RAM slots occupied (quad-channel mode) and RAM is DDR3 1600 MHz 9-9-9-27-2T and running at that rated speed and timings. This is from memory since I'm in linux and would have to boot into bios to confirm.
|
|
|
|
|
|
With only 1 day and 7 hours to go I have already stopped any more 321 LLR work units downloading.
Checking time to go and time it takes for me to do a WU I wont get all I have on the computers done by finish time.
So instead of aborting work I will just finish off what I have left and have already started working on some Genefer WUs (CPU type) as I found I don't have this badge yet, plus getting all bronze badges to silver.
Each Genefer work unit downloads with an estimated time of 1,000 hours but should finish in under 50 hours I think, maybe even under 40 hours, we shall see how they go.
Good luck to all fighting for positions, I am very happy with how I went and should finish well within the top 150 as an individual and within the top 50 for teams (I am the only one doing this challenge). Very pleased with that performance.
Still no primes but we keep searching don't we?
Conan
____________
|
|
|
|
|
|
Hi Conan,
Good challenge for you, have you realised that Aus has 4 teams in the top 100! Not too bad with under 10% of the US population.
I am doing similar to you with one PC, and leaving the other 2 on cleanup, going towards the gold on 321, maybe.
See you next challenge.
____________
Member team AUSTRALIA
My lucky number is 9291*2^1085585+1 |
|
|
|
|
Yeah, HT is definitely not worth it. I was thinking the huge L3 cache may have helped with that, but guess not for LLR. What clock speed is yours set at? Mine is currently at 4GHz and I am just barely under 30K seconds per WU.
3.8 GHz w/ all 4 RAM slots occupied (quad-channel mode) and RAM is DDR3 1600 MHz 9-9-9-27-2T and running at that rated speed and timings. This is from memory since I'm in linux and would have to boot into bios to confirm.
Hmm, I have all 8 slots occupied with quad channel memory, but I think its speed is at 1333 MHz. I should try increasing that.
____________
|
|
|
|
|
|
Qazwaplol also increased his memory from 1333MHz to 1866MHz, which made a 2 hour difference per task !!
In his post he said:
321LLR on a Intel Core i5 2500K, @ 4.5GHz:
- with 1333MHz RAM it takes around 9,5 hours per task
- with 1866MHz RAM it takes around 7,5 hours per task
so confirming what other already said in this thread.. RAM-speed does affect performance big time in this challenge :P
____________
Member of the Dutch Power Cows
My Stats |
|
|
|
|
Hi Conan,
Good challenge for you, have you realised that Aus has 4 teams in the top 100! Not too bad with under 10% of the US population.
I am doing similar to you with one PC, and leaving the other 2 on cleanup, going towards the gold on 321, maybe.
See you next challenge.
G'Day DaveB,
Yep both of us are doing quite well for with what we have.
I was keeping pace with BOINC@AUSTRALIA for a long time but they increased their output and left me behind.
However I have been having a great battle with the following teams as this snap shot shows, positions have changed numerous times.
33 L'Alliance Francophone______ 3005456.05
34 University of California, Berkeley 2984647.08
35 Chanology________________ 2840481.79
36 Cobar Spiders_____________ 2786621.66
37 UK BOINC Team___________ 2569960.55
38 Electronic Sports League (ESL) 2563544.86
So a good chance I will see you next challenge.
Catcha later
Conan
____________
|
|
|
|
|
Qazwaplol also increased his memory from 1333MHz to 1866MHz, which made a 2 hour difference per task !!
In his post he said:
321LLR on a Intel Core i5 2500K, @ 4.5GHz:
- with 1333MHz RAM it takes around 9,5 hours per task
- with 1866MHz RAM it takes around 7,5 hours per task
so confirming what other already said in this thread.. RAM-speed does affect performance big time in this challenge :P
Exciting, I will have to play around more once the challenge is over.
____________
|
|
|
|
|
|
I had noticed that my Linux machines, which I have added Genefer work to, all the 321 work units had slowed down to the point where I wont meet the deadline for return for the challenge.
This is disappointing as I should of been able to return at least 3-4 work units per machine (6-8 total) within time.
I had suspended the Genefer tasks so I could finish the 321 tasks.
I should of checked System Monitor earlier as the Genefer task did not suspend but kept running even though BOINC Manager said they were suspended.
This happened on both machines, so that the CPU was shared with twice the normal number of tasks (8 on the 4 core and 12 on the 6 core).
This slowed everything down but as I had to leave I did not investigate.
On my return I Stopped and restarted the BOINC Client and this fixed the problem and the suspended Genefer tasks are now not running any more.
I will probably loose a couple of placings due to this issue but thems the brakes I suppose.
It didn't happen on the Windows machines as they are 32 bit and can't run Genefer tasks.
Conan
____________
|
|
|
|
|
Yeah, HT is definitely not worth it. I was thinking the huge L3 cache may have helped with that, but guess not for LLR. What clock speed is yours set at? Mine is currently at 4GHz and I am just barely under 30K seconds per WU.
3.8 GHz w/ all 4 RAM slots occupied (quad-channel mode) and RAM is DDR3 1600 MHz 9-9-9-27-2T and running at that rated speed and timings. This is from memory since I'm in linux and would have to boot into bios to confirm.
Hmm, I have all 8 slots occupied with quad channel memory, but I think its speed is at 1333 MHz. I should try increasing that.
My mistake, I have all 8 slots occupied too. So memory speed is important for this project. Would there be a difference in quad channel vs dual channel? |
|
|
|
|
|
Congrats to Sicituradastra and the Cows. Very many thanks to PrimeSearchTeam for giving us a good old run and to all who took part and who made it possible.
____________
Oh Bondage? Up Yours.
http://www.youtube.com/watch?v=ogypBUCb7DA
 |
|
|
|
|
|
Thanks :) The roasted rodents don't even taste that bad, as long as you add enough sauce :) :)
____________
|
|
|
|
|
|
How many workunits need to be completed before the standings are final?
How many workunits were completed during the challenge?
How much did we increase n during the challenge?
____________
|
|
|
|
|
