| Author |
Message |
|---|
 Michael Goetz Volunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
                           
|
|
Tour de Primes 2020
Welcome to the 12th annual Tour de Primes. 2 is the first prime number...and the only even one. This makes it unique among prime numbers. Therefore, February is declared Prime month...being the 2nd month of the year. :) And there's no better way to pay homage to a prime number than to go out and find one. :) More precisely, a Top 5000 prime.
For the month of February, an informal competition is offered. There are no challenge points to be gained... just a simple rare jersey at the end of the month to add to your badge list. No pressure or stress other than what you put on yourself. :)
For 2020, we're bringing back the badges introduced in 2018, but since 2020 is a leap year, we're going to run the Mountain Stage competition on Leap Day, February 29th:
 Red Jersey -- discoverer of largest prime
 Yellow Jersey -- prime count leader (tiebreaker will be prime score)
 Green Jersey -- points (prime score) leader
 Polk-a-dot Jersey -- on the 29th (Leap Day) of February we'll have a "Mountain Stage" and award the Polk-a-dot Jersey to the one who finds the most primes on that day (tiebreaker will be prime score for that day).
 Prime badge -- awarded to everyone who finds an eligible prime during the month of February. This is a counter badge, so if you find more than one prime it will show how many you've found, up to 99.
 Mega prime badge -- awarded to everyone who finds a mega prime during February. This is a counter badge.
 Mountain Stage prime badge -- awarded to everyone who finds an eligible prime during the Mountain Stage. This is a counter badge.
 Mountain Stage mega prime badge -- awarded to everyone who finds a mega prime during the Mountain Stage. This is a counter badge.
Results will be available at http://www.primegrid.com/challenge/tdp_2020.php.
As we've done for the last couple of years, there will also be a "just for giggles" Double Checker scoreboard at http://www.primegrid.com/challenge/tdp_2020_dc.php.
As with the last few years, for all primes (BOINC and PRPNet) we're using the new reporting system whereby the prime's date of discovery determines whether it's eligible for the Tour de Primes. Prior to 2014, the date of verification for BOINC primes was used while the discovery date was used for PRPNet primes. The current system is more intuitive and fairer.
Note that SGS-LLR and GFN-15 are too small to be reported to the Top 5000 primes list and are therefore not eligible for the 2020 Tour de Primes.
Currently, the fastest opportunities to find Top 5000 primes is with the PPSE (LLR), and GFN-16 (65536) projects. Of course, should someone find a prime in the mega-prime searches, this would certainly give them a good shot at the green or red jerseys. Not a guarantee, however, as in 2019 there were almost 20 mega primes found in the Tour de Primes. Overall, in 2019 we averaged over two mega primes per week for the entire year, so you might need more than "merely" a mega prime to take home green.
All ports in PRPNet are available for the competition.
To participate in BOINC PPSE (LLR), GFN-16, or any other eligible LLR or Genefer project, all you have to do is select it in your PrimeGrid preferences. AP27 sequences are not reportable at T5K, so are not eligible for Tour de Primes.
Sieve projects don't find primes, so 321-sieve, PPS-Sieve, and manual sieving can't count in TdP.
Good Luck, have fun, and enjoy! :D
Previous Winners
 - Honza
 - Scott Brown
 - 288larsson
 - [AF>EDLS]GuL
 - Robish
 - [AF>Amis des Lapins] dthonon
 - [AF>Amis des Lapins] dthonon
 - tng*
 - wdethomas
 - tng*
 - wdethomas
 - tng*
 - boss
 - Scott Brown
 - boss
 - Orange_1050
 - Scott Brown
 - Randall J. Scalise
 - vmc
 - [DPC]x-RaY99_the_one_man_team
 - [DPC]x-RaY99_the_one_man_team
 - Lonnie Christensen
 - [DPC]x-RaY99_the_one_man_team
 - [DPC]x-RaY99_the_one_man_team
 - PBT_marian_boss
 - Scott Brown
 - Usucapio Libertatis
 - Scott Brown
 - shanky123
 - [BOINCstats] LostBoy
 - Snf*
 - lennart
 - lennart
 - [SG]marodeur6
 - lennart
 - s-yama
 - lennart
Full rankings can be seen here: 2009 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019
Totals by Year
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
Total Primes found 212 309 766 646 238 254 254 346 246 256 277
Total Score 457.30 2663.27 21803.72 20727.04 15614.83 20982.77 30268.46 43656.50 40201.94 55393.50 63588.96
Tips and Strategies:
Tip #1: He (or she) who finds the prime FIRST is the discover of the prime. It's a competition between you and your wingman. While having a fast computer helps, your computer is only useful when it's running a task. If you have a cache of tasks sitting on your computer waiting to run, chances are your wingman will return the task before you've even started it. Setting both BOINC cache settings to "0 days" is strongly recommended. People with slow computers find primes all the time because their wingman downloaded the task yesterday but won't start running it until tomorrow. Set your cache to 0 days!
If you're going to ignore tip #1, don't even bother reading the rest. Seriously.
Tips for LLR:
- Your mileage may vary. What works for me may not work for you. Before TdP starts, take some time and experiment and see what works best on your computer.
- If you have an Intel CPU with hyperthreading, either turn off the hyperthreading in the BIOS, or set BOINC to use 50% of the processors. (But see below for exceptions.)
- If you're using a GPU for other tasks, it may be beneficial to leave hyperthreading on in the BIOS and instead tell BOINC to use 50% of the CPU's. This will allow one of the hyperthreads to service the GPU.
- Use LLR's multithreaded mode. It requires a little bit of setup, but it's worth the effort. Follow these steps:
- Use multi-threading on larger tasks. What constitutes "larger" varies according to how fast your CPU is and your personal tastes.
- It's no longer necessary to use app_config.xml to set up multi-threading. On the PrimeGrid preferences page, set "Multi-threading: Max # of threads for each task" to "No limit" to use the maximum number of threads for each task, or select a specific thread count if you prefer.
Tips for GFN:
- Only run GFN on a GPU. Use your CPU for LLR tasks where it will be much more efficient.
- Unless you have a really slow GPU and a really fast CPU, leave a CPU core free to service the GPU. You'll want the more powerful GPU running at full speed, even if it slows down the CPU somewhat. A hyperthread should be sufficient if your CPU supports hyperthreading. For example, on a 4 core CPU (without hyperthreading), you could set BOINC to "use 75% of the CPUs" to reserve one core for the GPU.
____________
My lucky number is 75898524288+1
|
|
|
|
|
|
Finding a prime (returning the result) in Februrary is enough to count fo TdP ?
I mean, does the task have to be downloaded in February or may it be, for instance, downloaded January 31st and returned (with a prime found) on Feb 2nd ?
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment. |
|
|
tngSend message
Joined: 29 Aug 10
Posts: 398
ID: 66603
Credit: 22,925,088,044
RAC: 2
                                   
|
Finding a prime (returning the result) in Februrary is enough to count fo TdP ?
I mean, does the task have to be downloaded in February or may it be, for instance, downloaded January 31st and returned (with a prime found) on Feb 2nd ?
It's when the result is returned that matters for the TdP.
____________
 |
|
|
Dave
Send message
Joined: 13 Feb 12
Posts: 2829
ID: 130544
Credit: 954,793,678
RAC: 0
                
|
|
This is why I started a DYFL early. Mhwahaha. |
|
|
|
|
|
You know there's such a small chance of finding a prime ;) 🤡🤡
And you could run a GFN 17 and suspend it until 00:00UTC...
But these ways won't be effective unless you know it's gonna yield a prime, anyways I'm sticking to PPSE. ðŸ˜ðŸ˜
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
You know there's such a small chance of finding a prime ;) 🤡🤡
And you could run a GFN 17 and suspend it until 00:00UTC...
But these ways won't be effective unless you know it's gonna yield a prime, anyways I'm sticking to PPSE. ðŸ˜ðŸ˜
Actually, the problem is that if you do that, you're likely to be the double checker of a prime found before TdP rather than the finder of a TdP prime.
____________
My lucky number is 75898524288+1 |
|
|
|
|
You know there's such a small chance of finding a prime ;) 🤡🤡
And you could run a GFN 17 and suspend it until 00:00UTC...
But these ways won't be effective unless you know it's gonna yield a prime, anyways I'm sticking to PPSE. ðŸ˜ðŸ˜
Actually, the problem is that if you do that, you're likely to be the double checker of a prime found before TdP rather than the finder of a TdP prime.
It's an interesting way to game the double checker leader board.
____________
 |
|
|
|
|
It's an interesting way to game the double checker leader board.
My question is why would you want to do that? I'd rather be the initial finder :) |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
You know there's such a small chance of finding a prime ;) 🤡🤡
And you could run a GFN 17 and suspend it until 00:00UTC...
But these ways won't be effective unless you know it's gonna yield a prime, anyways I'm sticking to PPSE. ðŸ˜ðŸ˜
Actually, the problem is that if you do that, you're likely to be the double checker of a prime found before TdP rather than the finder of a TdP prime.
It's an interesting way to game the double checker leader board.
It’s more than fine with me if everyone else tries to be the double checker!!!
____________
My lucky number is 75898524288+1 |
|
|
|
|
You know there's such a small chance of finding a prime ;) 🤡🤡
And you could run a GFN 17 and suspend it until 00:00UTC...
But these ways won't be effective unless you know it's gonna yield a prime, anyways I'm sticking to PPSE. ðŸ˜ðŸ˜
Actually, the problem is that if you do that, you're likely to be the double checker of a prime found before TdP rather than the finder of a TdP prime.
It's an interesting way to game the double checker leader board.
It’s more than fine with me if everyone else tries to be the double checker!!!
The more, the better! :-) |
|
|
|
|
|
😂😂Yeah, and hope that they'd have a 10+10 cache...😂ðŸ˜
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
Sysadm@NbgVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 5 Feb 08
Posts: 1188
ID: 18646
Credit: 490,016,651
RAC: 0
            
|
|
while the preparations two prime numbers found, I predict I will find none in February during the tour
____________
Sysadm@Nbg
my current lucky number: 3749*2^1555697+1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/
 |
|
|
|
|
while the preparations two prime numbers found, I predict I will find none in February during the tour
I predict one per day at least!
____________
92*10^1439761-1 REPDIGIT PRIME :) :) :)
314187728^131072+1 GENERALIZED FERMAT
31*332^367560+1 CRUS PRIME
Proud member of team Aggie The Pew. Go Aggie! |
|
|
|
|
|
Last time I crunched PG was during last years TDP. In that time the old app_config stuff has changed. Can someone please point me in the direction on how to use the new set up in the options?
I know it's talked about here somewhere on the forum when the app_config changed but I can't find the thread. I'd rather share my confusion at setting up PG in that thread.
Thank you
____________
 |
|
|
|
|
|
You need an additional <plan_class>mt</plan_class>
Example:
<app>
<name>llrDIV</name>
<fraction_done_exact/>
<max_concurrent>2</max_concurrent>
</app>
<app_version>
<app_name>llrDIV</app_name>
<plan_class>mt</plan_class>
<cmdline>-t 6</cmdline>
<avg_ncpus>1</avg_ncpus>
<max_ncpus>1</max_ncpus>
</app_version>
____________
DeleteNull |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
Last time I crunched PG was during last years TDP. In that time the old app_config stuff has changed. Can someone please point me in the direction on how to use the new set up in the options?
I know it's talked about here somewhere on the forum when the app_config changed but I can't find the thread. I'd rather share my confusion at setting up PG in that thread.
Thank you
In this forum, there's a sticky thread titled "Better Multi-Threading". That's what you want.
The short version is 1) delete app_config.xml and then 2) use the new control on the preferences web page where you select which tasks you want to run.
There's two new controls:
Max # of simultaneous PrimeGrid tasks: Leave this set to "No limit".
Multi-threading: Max # of threads for each task: This conrtols how many threads your LLR tasks will each use. (Only LLR tasks are affected.) It defaults to 1, which means the tasks don't use multi-threading. Set it to 2, and each task will use 2 threads. Set it to "No limit" and BOINC will run a single task that uses all the available threads.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Thank you.
I used this file last night which I think is correct for my threadripper to run 1 w/u across 16 cores with CPU usage set to 50%
<app>
<app_name>llrPPSE</app_name>
<max_concurrent>1</max_concurrent>
</app>
<app_version>
<app_name>llrPPSE</app_name>
<plan_class>mt</plan_class>
<cmdline>-t 16</cmdline>
<avg_ncpus>16</avg_ncpus>
<max_ncpus>16</max_ncpus>
</app_version>
I'll ask my questions on the Better Multithreading post.
Kind regards
____________
|
|
|
|
|
|
I started PPSE but something interesting happens. My computer has 6 cores with 12 threads. While computing only 1 tasks using 1 thread , it took 13 min to finish this taks. But, if computing 4 tasks using 2 threads each (totally 8 threads), it takes 20 min per tasks. How could this circumstance happen? :( |
|
|
pewSend message
Joined: 24 Jan 16
Posts: 2
ID: 437165
Credit: 175,450,520
RAC: 0
         
|
|
It's most likely because of hyperthreading. If you ran 3 tasks with 2 threads, i.e. for 6 cores, you might see an improvement in time. When llr has to run on both logical threads of a core it's much less efficient. Usually you want to either use only 50% of cpu or turn off hyperthreading when doing llr tasks. Also, going from 1 to 2 cores for ppse, and any 'small' tasks, usually doesn't increase the throughput in any good way. Hopefully that makes sense. |
|
|
robishVolunteer moderator
Volunteer tester
Send message
Joined: 7 Jan 12
Posts: 1774
ID: 126266
Credit: 5,066,569,935
RAC: 0
            
|
It's most likely because of hyperthreading. If you ran 3 tasks with 2 threads, i.e. for 6 cores, you might see an improvement in time. When llr has to run on both logical threads of a core it's much less efficient. Usually you want to either use only 50% of cpu or turn off hyperthreading when doing llr tasks. Also, going from 1 to 2 cores for ppse, and any 'small' tasks, usually doesn't increase the throughput in any good way. Hopefully that makes sense.
ðŸ‘
____________
My lucky numbers 10590941048576+1 and 224584605939537911+81292139*23#*n for n=0..26 |
|
|
|
|
|
It exactly makes sense! Thanks a lot! |
|
|
|
|
|
And so it begins. This is the first year since the new badges were being handed out that I've been able to participate, I'm really hoping to strike a prime. I'm still really disappointed my participation in early years did not result in retroactively awarded badges, but alas. This year will be the year.
____________
|
|
|
|
|
And so it begins. This is the first year since the new badges were being handed out that I've been able to participate, I'm really hoping to strike a prime. I'm still really disappointed my participation in early years did not result in retroactively awarded badges, but alas. This year will be the year.
And best of luck. I am sure you will!
Cheers |
|
|
|
|
|
BOOM! Renix1943 with the first! Congratulations! And you too double checker! |
|
|
|
|
|
Congrats Renix!!
____________
|
|
|
|
|
Congrats Renix!!
Thank you :-)
|
|
|
|
|
|
It seems that we got someone earlier :)
And IDK how he did this double kill :( |
|
|
|
|
And IDK how he did this double kill :(
he/she ;)
He/she got the next prime too.[/quote]
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
|
|
|
Wow, we had 2 PPS-MEGA discovered last night.
____________
My DC mathematical side :)
 |
|
|
|
|
|
Well, one day done, and I double checked a result. Hopefully I continue to have more luck. Probability says that I will only run two tasks for primes during the month, so we will see how things go.
____________
|
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
      
|
Well, one day done, and I double checked a result. Hopefully I continue to have more luck. Probability says that I will only run two tasks for primes during the month, so we will see how things go.
Finding a prime doesn't count against you, statistics-wise. So probability still says you will run about two more tasks for primes during the month. |
|
|
|
|
|
The server keeps sending me four tasks at once even though I have both the fetch/store options set to 0 (and have done for like, over a year). I'm not going to find anything. :(
____________
 |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
The server keeps sending me four tasks at once even though I have both the fetch/store options set to 0 (and have done for like, over a year). I'm not going to find anything. :(
In an app_config.xml, you can set <avg_ncpus>1</avg_ncpus>, no matter what the -t setting is. Lower your % CPU used in BOINC to match. Then BOINC doesn't know it's not running single-threaded, but LLR knows to run multiple threads. |
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 882
ID: 105020
Credit: 1,318,826,036
RAC: 0
             
|
The server keeps sending me four tasks at once even though I have both the fetch/store options set to 0 (and have done for like, over a year). I'm not going to find anything. :(
I wonder if setting max # of simultaneous PG tasks to the number of tasks you are doing may help (rather than no limit)?
I know that has the potential to go hilariously wrong if you are running CPU and GPU tasks - it looks like you are only running CPU tasks, so that particular problem won't occur. |
|
|
|
|
Well, one day done, and I double checked a result. Hopefully I continue to have more luck. Probability says that I will only run two tasks for primes during the month, so we will see how things go.
Finding a prime doesn't count against you, statistics-wise. So probability still says you will run about two more tasks for primes during the month.
You must resource share to 0 in Primegrid project. That is very important
____________
92*10^1439761-1 REPDIGIT PRIME :) :) :)
314187728^131072+1 GENERALIZED FERMAT
31*332^367560+1 CRUS PRIME
Proud member of team Aggie The Pew. Go Aggie! |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
|
I did a rough statistical analysis of 197 PPSE WUs one of my machines acquired, to figure out how completion time relates to your likelihood of being 1st. I compared the time my computer received the work to the time the other computer reported the work, so my computer's speed doesn't matter. If I'd checked what error bars that would produce first, I might have collected more WUs, but oh well.
If you complete within (s), then you can expect to be first (+/- 7%)
222 99%
236 98%
242 97%
249 96%
297 95%
330 94%
334 93%
342 92%
357 91%
365 90%
369 89%
380 88%
394 87%
409 86%
427 85%
433 84%
449 83%
452 82%
460 81%
469 80%
473 79%
485 78%
492 77%
500 76%
508 75%
515 74%
571 73%
587 72%
594 71%
597 70%
602 69%
611 68%
614 67%
625 66%
630 65%
636 64%
642 63%
652 62%
661 61%
665 60%
667 59%
681 58%
683 57%
689 56%
696 55%
706 54%
709 53%
719 52%
734 51%
747 50%
751 49%
755 48%
763 47%
772 46%
777 45%
779 44%
795 43%
798 42%
802 41%
811 40%
821 39%
823 38%
839 37%
846 36%
854 35%
860 34%
862 33%
865 32%
875 31%
884 30%
891 29%
899 28%
920 27%
923 26%
953 25%
961 24%
1026 23%
1043 22%
1068 21%
1079 20%
1105 19%
1152 18%
1165 17%
1170 16%
1201 15%
1212 14%
1233 13%
1284 12%
1305 11%
1311 10%
1523 9%
1584 8%
1678 7%
1886 6%
2303 5%
2434 4%
2486 3%
5625 2%
20053 1%
53294 0%
|
|
|
|
|
|
Some numbers I worked out for my systems earlier. For each system, I just looked at 1st %, and average unit time.
41% 849
47% 746
58% 631
66% 614
85% 446
From above list, matching to nearest time:
846 36%
747 50%
630 65%
614 67%
449 83%
Think we're not that far apart. If people start changing behaviour to increase their 1st chances, then this wont remain constant over time. |
|
|
|
|
I did a rough statistical analysis of 197 PPSE WUs one of my machines acquired, to figure out how completion time relates to your likelihood of being 1st. I compared the time my computer received the work to the time the other computer reported the work, so my computer's speed doesn't matter. If I'd checked what error bars that would produce first, I might have collected more WUs, but oh well.
If you complete within (s), then you can expect to be first (+/- 7%)
222 99%
236 98%
242 97%
249 96%
297 95%
330 94%
334 93%
342 92%
357 91%
365 90%
369 89%
380 88%
394 87%
409 86%
427 85%
433 84%
449 83%
452 82%
460 81%
469 80%
473 79%
485 78%
492 77%
500 76%
508 75%
515 74%
571 73%
587 72%
594 71%
597 70%
602 69%
611 68%
614 67%
625 66%
630 65%
636 64%
642 63%
652 62%
661 61%
665 60%
667 59%
681 58%
683 57%
689 56%
696 55%
706 54%
709 53%
719 52%
734 51%
747 50%
751 49%
755 48%
763 47%
772 46%
777 45%
779 44%
795 43%
798 42%
802 41%
811 40%
821 39%
823 38%
839 37%
846 36%
854 35%
860 34%
862 33%
865 32%
875 31%
884 30%
891 29%
899 28%
920 27%
923 26%
953 25%
961 24%
1026 23%
1043 22%
1068 21%
1079 20%
1105 19%
1152 18%
1165 17%
1170 16%
1201 15%
1212 14%
1233 13%
1284 12%
1305 11%
1311 10%
1523 9%
1584 8%
1678 7%
1886 6%
2303 5%
2434 4%
2486 3%
5625 2%
20053 1%
53294 0%
Interesting. Thanks!
I briefly tried running 8 tasks with 2 cores on each with my 1950x but my times blew out into the 900’s and there were very few firsts, so I changed back to what I had before being 4*4.
____________
|
|
|
|
|
|
I didn't expect to get the TdP badge since it's my first time participating in it. But luck came last night and I appreciated. It would be the best present for one-year anniversary for my joining PG family.
Too excited to fall asleep
Yes, I feel lucky today so I will start a DYFL task :) |
|
|
|
|
I didn't expect to get the TdP badge since it's my first time participating in it. But luck came last night and I appreciated. It would be the best present for one-year anniversary for my joining PG family.
Too excited to fall asleep
Yes, I feel lucky today so I will start a DYFL task :)
btw, to increase the chance of being the initial finder of GFN-16 tasks, “NOT TO accumulate†is the trick. 50% of your task will be "completed(1st)" if you can finish a GNF-16 tasks within 4 minutes (the faster, the better) and set "NOT TO accumulate". You can set your PG preference: only 1 task simultaneously, to avoid tasks from accumulating.
To get 1st, time matters. The whole period of one task is composed of:
1.Time for downloading
2.Time for waiting for the last task to be finished
3.Time for calculating
4.Time for uploading and reporting
The second one, which may double the whole process, can be avoided by following the aforementioned method.
Good luck guys! |
|
|
|
|
btw, to increase the chance of being the initial finder of GFN-16 tasks, “NOT TO accumulate†is the trick. 50% of your task will be "completed(1st)" if you can finish a GNF-16 tasks within 4 minutes (the faster, the better) and set "NOT TO accumulate". You can set your PG preference: only 1 task simultaneously, to avoid tasks from accumulating.
To get 1st, time matters. The whole period of one task is composed of:
1.Time for downloading
2.Time for waiting for the last task to be finished
3.Time for calculating
4.Time for uploading and reporting
The second one, which may double the whole process, can be avoided by following the aforementioned method.
Good luck guys!
The problem is, my average GFN16 task takes 25 minutes to complete....
I have minimized 1 and 4. And I am already using 0+0cache. But my 3 just isn't fast enough...
In fact, it's so slow that I've been double checking all results since TdP started--no 1sts.
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
|
|
The server keeps sending me four tasks at once even though I have both the fetch/store options set to 0 (and have done for like, over a year). I'm not going to find anything. :(
I wonder if setting max # of simultaneous PG tasks to the number of tasks you are doing may help (rather than no limit)?
I know that has the potential to go hilariously wrong if you are running CPU and GPU tasks - it looks like you are only running CPU tasks, so that particular problem won't occur.
Thanks, Nick, that fixed it. The odd thing is that it would alternate between sending me four tasks for a few hours, and then go to one for a while, and then repeat. Re the GPU situation, I was about to replace my dead unit when nature intervened and I decided to reallocate that money for the meantime in case things get bad.
____________
|
|
|
|
|
I didn't expect to get the TdP badge since it's my first time participating in it. But luck came last night and I appreciated. It would be the best present for one-year anniversary for my joining PG family.
Too excited to fall asleep
Yes, I feel lucky today so I will start a DYFL task :)
Congratulations!
____________
|
|
|
|
|
|
Well, 25 minutes seems to be too long to be initial finder.....
You're using AMD Vega8, similar to Nvidia 650. You can check the "Fastest GPU list" on the PG website to check its normal average speed recorded by PG users (My GTX1060 takes 4 minutes and RTX2060 takes 1.5 minutes for GFN16, which you can use as references to calculate your actual time of processing one task with AMD). Particularly, if your AMD is obviously slower than what the list shows, check the cooling system and rate of utilization in the task manager.
If your AMD v8 is exactly running with its best capability but takes 25 minutes for one GFN16 task, I suggest that you can switch to PPSE or PPS-MEGA (Your CPUs are better) since there's little chance for AMD V8 to get "completed 1st" while treasures in PPSE and MEGA await :)
Good luck! |
|
|
|
|
|
Using proportions, I found that with your 2060 as 1.5 minutes, my V8 is exactly on par--25.01 is the average. Or perhaps I'm the only V8?
I am using 3 cores on PPSE and leaving a core for GFN16. I am also getting 17low, just for fun cuz I know I'm not gonna get a prime.
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
|
|
|
Hmm I'm not sure I'd bother with that Vega GPU.
You certainly had some patience with that GFN21!!! 1.22 Million seconds to complete.
I agree with ZephyR. You're probably better donating all cores to PPSE or another CPU task.
Goodluck!
____________
|
|
|
|
|
|
Hi, my host 2990wx dont download cpu tasks, onlu gpu, in this conifg
Max # of simultaneous PrimeGrid tasks 7
Multi-threading: Max # of threads for each task 8
I do many reset becouse , when i change that numbers , boinc manager dont react on change.. i was try stop project, refresh,readconfig ..etc.. only after reseting project happen change from 8-8, max-8, 7-7, or other what i trying before...
dont use any xml or config.
know somebody what to do now?
thank
|
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 882
ID: 105020
Credit: 1,318,826,036
RAC: 0
|
Hi, my host 2990wx dont download cpu tasks, onlu gpu, in this conifg
Max # of simultaneous PrimeGrid tasks 7
Multi-threading: Max # of threads for each task 8
I do many reset becouse , when i change that numbers , boinc manager dont react on change.. i was try stop project, refresh,readconfig ..etc.. only after reseting project happen change from 8-8, max-8, 7-7, or other what i trying before...
dont use any xml or config.
know somebody what to do now?
thank
You may have a couple of issues.
I think I know one of them.
If you are running CPU & GPU tasks - when you specify max # simultaneous PG tasks, PG can reach a scenario where it isn't downloading GPU tasks - maybe it could also not be downloading CPU tasks which is probably less likely.
If doing CPU and GPU tasks, another way to speed things up is to set resource share to zero in PG preferences. (and set Max # simultaneous tasks to no limit) This should have the outcome of downloading the next task when the previous is finished. I think setting resource share to zero only works if PG is the only project running.
For anyone running only CPU tasks OR only GPU tasks, I understand setting max # simultaneous tasks to the number you want running at once (or the number of tasks that can run at once) will stop tasks downloading and sitting there waiting to run - they will download once a task is finished. |
|
|
|
|
You certainly had some patience with that GFN21!!! 1.22 Million seconds to complete.
that was on CPU ;)
I tried in the summer with GPU but it always had an error when I close the computer for the night.
Well I had patience too with anotehr SoB task which had an error at 98% and had been running for 5 days!!!! Imagine how mad I was at my laptop.
I do many reset becouse , when i change that numbers , boinc manager dont react on change.. i was try stop project, refresh,readconfig ..etc.. only after reseting project happen change from 8-8, max-8, 7-7, or other what i trying before...
Maybe updating the project will help..? Since stopping and reading config doesn't tell the manager to contact the PG server, only updatting or resetting could help, and updating is a lot faster.
Max # of simultaneous PrimeGrid tasks 7
Set this to unlimited first and see what happens, possibly its downloading GPU tasks all the time?
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
|
|
You certainly had some patience with that GFN21!!! 1.22 Million seconds to complete.
that was on CPU ;)
I tried in the summer with GPU but it always had an error when I close the computer for the night.
Well I had patience too with anotehr SoB task which had an error at 98% and had been running for 5 days!!!! Imagine how mad I was at my laptop.
Ah ok.
LOL...yes I think I lost a SoB wu once. Can't afford that to happen too much with electricity prices here in Australia :(
Tough time of year for hard crunching too. We've had temps up around 38 Celsius (100F) at times recently.
Can't wait for winter!
____________
|
|
|
|
|
|
I gotta tell y'all I have been so hyped for the Tour de Primes. I was super disappointed I only got a double check last year, but that machine was slow as all getout. My new machine is plowing through the PPSEs like nobody's business so I think I got a shot. I can't wait for that sweet, sweet taste of prime discovery, I really think it might happen this year.
I... don't have a lot else going on in my life. |
|
|
|
|
I gotta tell y'all I have been so hyped for the Tour de Primes. I was super disappointed I only got a double check last year, but that machine was slow as all getout. My new machine is plowing through the PPSEs like nobody's business so I think I got a shot. I can't wait for that sweet, sweet taste of prime discovery, I really think it might happen this year.
I... don't have a lot else going on in my life.
Good luck Zach I sure hope you find a prime, it is a good feeling.
I think one is heading your way, eep on crunching.
PPSE and GFN16 are your best bet
|
|
|
|
|
I gotta tell y'all I have been so hyped for the Tour de Primes. I was super disappointed I only got a double check last year, but that machine was slow as all getout. My new machine is plowing through the PPSEs like nobody's business so I think I got a shot. I can't wait for that sweet, sweet taste of prime discovery, I really think it might happen this year.
I... don't have a lot else going on in my life.
I just had a look and you are getting a lot of 1st in PPSE mate
I reckon you have agreat chance this year, so keep persisting and best wishes:) |
|
|
|
|
|
Good luck and best wishes!
I'm planning to switch to 321 LLR now...
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
|
|
|
Got a PPSE prime!
Moving to larger projects now. Good luck everyone!
(MT is still worth on small tasks. Got mine 30 seconds after the doublechecker, returned it roughly 1 minute earlier)
____________
676754^262144+1 is prime |
|
|
|
|
(MT is still worth on small tasks. Got mine 30 seconds after the doublechecker, returned it roughly 1 minute earlier)
Agreed!!!!
If there was no MT I would never be 1st...
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
|
Pro tip: In terms of first, the metric everyone should be looking at is "1sts per hour". The raw percentage doesn't really matter. It's how many you get per hour that's important.
In many cases, that still means more threads are better, but this will show you when the diminishing returns are too severe.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Is 3T per PPSE too much for you?😂😂😂😂
15min/task, ~20% first. 1 first per hour.
Should I just abandon this TdP altogether and wait till I've got a better machine?
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
Dave
Send message
Joined: 13 Feb 12
Posts: 2829
ID: 130544
Credit: 954,793,678
RAC: 0
|
|
The moment you give up you'll wish you hadn't... |
|
|
|
|
Is 3T per PPSE too much for you?😂😂😂😂
15min/task, ~20% first. 1 first per hour.
Should I just abandon this TdP altogether and wait till I've got a better machine?
Disable SMT or use 50% threads (=4).
Run 1 task by 4 threads. ;)
Edit: I noticed you have 1 GPU. Mmm, it looks very very slow.
Disable GPU. :D
____________
My DC mathematical side :)
|
|
|
|
|
|
Is wing man selected by different computer or different user? Can you be your own wing man if you have several computers? |
|
|
|
|
Can you be your own wing man if you have several computers?
No, it happened some years ago.
____________
My DC mathematical side :)
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
Can you be your own wing man if you have several computers?
No.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Here it is: https://www.primegrid.com/forum_thread.php?id=7805&nowrap=true#114518
____________
My DC mathematical side :)
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
Here it is: https://www.primegrid.com/forum_thread.php?id=7805&nowrap=true#114518
This happens all the time with some of our heavy hitters. :)
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
After 10 years I finally found my first prime during Tour de Primes. Yippie!
____________
My lucky number: 65531*2^3629342-1 (TRP) |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
After 10 years I finally found my first prime during Tour de Primes. Yippie!
Congratulations!
____________
My lucky number is 75898524288+1 |
|
|
|
|
After 10 years I finally found my first prime during Tour de Primes. Yippie!
Woohoo !
Congrats !
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment. |
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 882
ID: 105020
Credit: 1,318,826,036
RAC: 0
|
After 10 years I finally found my first prime during Tour de Primes. Yippie!
That is a really long wait. Congratulations on breaking the drought! |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
Can you be your own wing man if you have several computers?
No.
Heh. Robish reminded me of a conversation from a few months back...
By design, this is prohibited, but there's a race condition where it can happen anyway.
If you have two computers A and B, and both A and B request work at exactly the same moment, when the server (which is multi-threaded and can literally be doing multiple things simultaneously) processes those two requests, this can happen:
1) Server gets request from A
2) Server gets request from B
3) Server finds task 0 in WU X for host A
4) Server finds task 1 in WU X for host B
5) Server checks in the database, for host A, if any tasks for WU X have already been sent to this user. None have been sent yet, so this task is permitted to be sent to host A.
6) Server checks in the database, for host B, if any tasks for WU X have already been sent to this user. None have been sent yet, so this task is permitted to be sent to host B.
7) Task 0 is sent to host A and records this in the database.
8) Task 1 is sent to host B and records this in the database.
The problem is that the checks are done against the database in steps 5 & 6, but the database isn't modified until after that happens, in steps 7 & 8.
If this happens, the validator will force a third task to be sent out so that there is, in fact tasks from two different users.
This doesn't happen very often.
____________
My lucky number is 75898524288+1 |
|
|
|
|
After 10 years I finally found my first prime during Tour de Primes. Yippie!
Congrats and well done!
Cheers Rick |
|
|
robishVolunteer moderator
Volunteer tester
Send message
Joined: 7 Jan 12
Posts: 1774
ID: 126266
Credit: 5,066,569,935
RAC: 0
|
|
Thanks Mike :)
Might as well share the link so as to clarify.
https://www.primegrid.com/workunit.php?wuid=640908453
Kellen got 1st and 2nd tasks at the exact same time. Kinda cool :)
Badge? :D
Edit: Worth a shot :D
____________
My lucky numbers 10590941048576+1 and 224584605939537911+81292139*23#*n for n=0..26 |
|
|
|
|
After 10 years I finally found my first prime during Tour de Primes. Yippie!
Congratulations, way to stick with it. |
|
|
|
|
After 10 years I finally found my first prime during Tour de Primes. Yippie!
Wow congrats!!
I reckon it would take me 10yrs or more to get a "K" or an "M" lol
____________
|
|
|
|
|
|
Sorry for the silly question but which has better odds of finding a Mega? PPS Mega or GFN 17 Mega?
I wouldn't mind an "M" :D
____________
|
|
|
|
|
Sorry for the silly question but which has better odds of finding a Mega? PPS Mega or GFN 17 Mega?
I wouldn't mind an "M" :D
I'm not sure about PPS Mega, but according to https://www.primegrid.com/stats_genefer.php there is a prime found on GFN-17 Mega once every 30 thousand workunits roughly.
I'm optimistically running GFN-17 Mega for Tour de Primes but at only 500 workunits per day, I've got a bit less than 50% chance of getting one this month.
Would be interested to know the stats for PPS Mega.
____________
 |
|
|
|
|
|
In January I made a thorough analysis of the subprojects and some interesting results turned out:
Finding a prime in SGS requires almost twice as many WUs as PPSE.
Finding a prime in GFN19 requires about 7000 less WUs than GFN18, probably due to the GFN18 drought last year.
Though Cullen numbers are smaller, they require more WUs to find a prime than Woodall.
It takes less WUs to find a SoB prime than a PSP.
(not that you shouldn't run PSP, cuz SoB is almost twice longer and therefore takes more time to find a prime, same for others above)
So it depends on what GPU/CPU you're throwing into the effort. 17MEGA is around 29560 and PPS-MEGA is around 35160.
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
|
|
In January I made a thorough analysis of the subprojects and some interesting results turned out:
Finding a prime in SGS requires almost twice as many WUs as PPSE.
Finding a prime in GFN19 requires about 7000 less WUs than GFN18, probably due to the GFN18 drought last year.
Though Cullen numbers are smaller, they require more WUs to find a prime than Woodall.
It takes less WUs to find a SoB prime than a PSP.
(not that you shouldn't run PSP, cuz SoB is almost twice longer and therefore takes more time to find a prime, same for others above)
So it depends on what GPU/CPU you're throwing into the effort. 17MEGA is around 29560 and PPS-MEGA is around 35160.
Cool thanks guys.
I'm handicapped in the GPU department. Running Radeon cards :D
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
Sorry for the silly question but which has better odds of finding a Mega? PPS Mega or GFN 17 Mega?
I wouldn't mind an "M" :D
That's a question you really don't need to be asking. I'll get to that in a moment.
The answer to your question is "about the same". The odds of finding a prime mostly depend on the size of the numbers. Since they're about the same size, the odds are about the same.
The reason I said you don't need to be asking this question is that one of those two (iGFN-17_MEGA) is a GPU project and the other (PPS-MEGA) is a CPU project. So run PPS-MEGA on your CPU and GFN-17-MEGA on your GPU.
Megas of any type are really hard to find.Good luck!
P.S. While dannyridel's analysis of the data is probably correct, his conclusion is wrong for several reasons. The bigger the numbers, the harder they are to find. A good rule of thumb is that everytime you double the number of digits, it takes 8 times as long to find a prime.
GFN-17's have about twice as many digits as GFN-16s...
____________
My lucky number is 75898524288+1 |
|
|
|
|
Can you be your own wing man if you have several computers?
... By design, this is prohibited, but there's a race condition where it can happen anyway. ...
Dijkstra famously published in 1965 the solution to race conditions in software.
But every single software designer using concurrency has to be aware of the solution, by reading about it, by being shown the solution, or by reinventing it. |
|
|
|
|
|
This morning I received an email re GFN Mega find, but it does not appear in the TdP or my primes list.
1061352756 974355 5 Feb 2020 | 19:35:55 UTC 5 Feb 2020 | 19:39:00 UTC Completed and validated (1st) 168.31 92.33 469.88 Genefer 17 Mega v3.21 (OCLcudaGFN17MEGA
Is this my impatience showing, or am I misinterpreting something.
Advice appreciated, and thank you for the TdP - really good crunching.
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 785
ID: 18447
Credit: 263,436,450
RAC: 0
    
|
|
It's probably awaiting test on either the PG and/or T5K server
____________
 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
It's probably awaiting test on either the PG and/or T5K server
At T5K.
We don't send the email until we submit to T5K, and we don't submit until our tests are done.
It will show up on PG within half an hour of it being verified on T5K.
____________
My lucky number is 75898524288+1 |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
Can you be your own wing man if you have several computers?
... By design, this is prohibited, but there's a race condition where it can happen anyway. ...
Dijkstra famously published in 1965 the solution to race conditions in software.
But every single software designer using concurrency has to be aware of the solution, by reading about it, by being shown the solution, or by reinventing it.
Sometimes solving a race condition is so resource-intensive (e.g. limiting multithreading) that it's better to deal with the occasional race condition. |
|
|
|
|
|
Thank you.
All posted now, and I've got my first Mega. :-)
Of all the projects I'v taken part in on Boinc et al, PG is OUTSTANDING in its ability to allow interaction with the database, and involvement in searching the data relationships etc.
I take my hat off to volunteers, and designers who have created and maintained it. Thanks.
____________
|
|
|
|
|
Sorry for the silly question but which has better odds of finding a Mega? PPS Mega or GFN 17 Mega?
I wouldn't mind an "M" :D
That's a question you really don't need to be asking. I'll get to that in a moment.
The answer to your question is "about the same". The odds of finding a prime mostly depend on the size of the numbers. Since they're about the same size, the odds are about the same.
The reason I said you don't need to be asking this question is that one of those two (iGFN-17_MEGA) is a GPU project and the other (PPS-MEGA) is a CPU project. So run PPS-MEGA on your CPU and GFN-17-MEGA on your GPU.
Megas of any type are really hard to find.Good luck!
P.S. While dannyridel's analysis of the data is probably correct, his conclusion is wrong for several reasons. The bigger the numbers, the harder they are to find. A good rule of thumb is that everytime you double the number of digits, it takes 8 times as long to find a prime.
GFN-17's have about twice as many digits as GFN-16s...
Hi Micheal,
Thank you for the reply. My aplogies, I didn't word it very well as I do know that one's a GPU vs CPU task. I just wasn't sure if the odds were different from running the GPU task or the CPU task.... but you've answered that for me...about the same :)
Found a GFN-16!!
Happy about that. One of the few with a Radeon card too :D
____________
|
|
|
|
|
Thank you.
All posted now, and I've got my first Mega. :-)
Of all the projects I'v taken part in on Boinc et al, PG is OUTSTANDING in its ability to allow interaction with the database, and involvement in searching the data relationships etc.
I take my hat off to volunteers, and designers who have created and maintained it. Thanks.
Agreed. This is why I donate directly here. This is easily one of the very best run projects in the BOINC universe - both past and present.
____________
 |
|
|
tngSend message
Joined: 29 Aug 10
Posts: 398
ID: 66603
Credit: 22,925,088,044
RAC: 2
|
Thank you.
All posted now, and I've got my first Mega. :-)
Of all the projects I'v taken part in on Boinc et al, PG is OUTSTANDING in its ability to allow interaction with the database, and involvement in searching the data relationships etc.
I take my hat off to volunteers, and designers who have created and maintained it. Thanks.
Agreed. This is why I donate directly here. This is easily one of the very best run projects in the BOINC universe - both past and present.
Same here -- well run, excellent communication. Many thanks to all involved.
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
Thank you.
All posted now, and I've got my first Mega. :-)
Of all the projects I'v taken part in on Boinc et al, PG is OUTSTANDING in its ability to allow interaction with the database, and involvement in searching the data relationships etc.
I take my hat off to volunteers, and designers who have created and maintained it. Thanks.
Thanks, the kind words are appreciated.
Conversely, we also have the best and craziest most dedicated users in the BOINC universe! :)
____________
My lucky number is 75898524288+1 |
|
|
|
|
Thanks, the kind words are appreciated.
Conversely, we also have the best and craziest most dedicated users in the BOINC universe! :)
I think that actually fits perfectly! We do have the best users and admins! And to think everyone is a volunteer. Making science happen all the time. |
|
|
|
|
|
Having issues getting PPSE (LLR) v8.04 (mt) Task and because of this I am getting 3 Tasks Genefer 16 v3.21 (OCLcudaGFN16) One Running and 2 Waiting to Run. I am NOT seeing this Issue on any other Host Running Windows 10.
Linux Ubuntu 18.04.4 LTS [4.15.0-87-generic|libc 2.27 (Ubuntu GLIBC 2.27-3ubuntu1)]
Intel(R) Core(TM) i7-3970X CPU @ 3.50GHz [Family 6 Model 45 Stepping 7]
NVIDIA GeForce GTX 1080 Ti (4095MB) driver: 44044
BOINC version 7.16.3
Host https://www.primegrid.com/show_host_detail.php?hostid=917582
Same issue running Genefer 17 Low v3.21 (OCLcudaGFN17LOW) But had to stop running this one because of so many errors from other members with Graphics Driver Issues.
So I can only run 1 GPU Task and No CPU Tasks, Kind of a Bummer.
PS my Settings are the Same on my Windows 10 Host and I am NOT having this same problem. |
|
|
KeithVolunteer tester
Send message
Joined: 8 Dec 13
Posts: 436
ID: 284516
Credit: 412,580,108
RAC: 0
       
|
Having issues getting PPSE (LLR) v8.04 (mt) Task and because of this I am getting 3 Tasks Genefer 16 v3.21 (OCLcudaGFN16) One Running and 2 Waiting to Run. I am NOT seeing this Issue on any other Host Running Windows 10.
If it's in a different venue than the other systems then make sure that venue has Resource Share set to 0 instead of the default of 100.
____________
My Primes
Badge Score: 2*1 + 3*1 + 4*2 + 6*4 + 7*9 + 9*1 + 10*2 = 129
 |
|
|
|
|
Thanks, the kind words are appreciated.
Conversely, we also have the best and craziest most dedicated users in the BOINC universe! :)
I think that actually fits perfectly! We do have the best users and admins! And to think everyone is a volunteer. Making science happen all the time.
Yes! Absolutely agree!!! "We're all math geeks, remember?"(Goetz 2014)
Though a sad thing is that most users are in US or Europe, when they're active I'm sleeping😂😂
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
|
|
Thanks, the kind words are appreciated.
Conversely, we also have the best and craziest most dedicated users in the BOINC universe! :)
I think that actually fits perfectly! We do have the best users and admins! And to think everyone is a volunteer. Making science happen all the time.
Yes! Absolutely agree!!! "We're all math geeks, remember?"(Goetz 2014)
Though a sad thing is that most users are in US or Europe, when they're active I'm sleeping😂😂
Hahaha it's the same situation with me, living with the Asian daily routine opposite to most PG users in the US and Europe.
My computer keeps running BOINC when I'm sleeping, especially during TdP. I can't play video games since my NVIDIA is working at 100% power with GFN tasks because I want to catch every chance to get the badge, which is annoying. My ranks in some computer games keep dropping these days. 😂 |
|
|
|
|
Having issues getting PPSE (LLR) v8.04 (mt) Task and because of this I am getting 3 Tasks Genefer 16 v3.21 (OCLcudaGFN16) One Running and 2 Waiting to Run. I am NOT seeing this Issue on any other Host Running Windows 10.
If it's in a different venue than the other systems then make sure that venue has Resource Share set to 0 instead of the default of 100.
Thank you,
All my settings are the same and all my settings are set to Zero (0) for Resource Share.
 |
|
|
|
|
|
Woot! That's 2 :D
Flyin the flag for AMD LOL!!
____________
|
|
|
|
|
Woot! That's 2 :D
Flyin the flag for AMD LOL!!
Rub it in !!!! Congrats on the Prims. Mine will come I hope.
____________
Proud Founder of 
Have a look at my WebCam
My best Prime 46776558131072+1 |
|
|
|
|
|
I don't have a MEGA though Dingo! :D
____________
|
|
|
|
|
|
Conversely, we also have the best and craziest most dedicated users in the BOINC universe! :)[/quote]
I think I've developed O C D, or obsessive Challenge disorder. Very worrying, causing excessive sweating over results, high temperatures on processing, and general desire to spend loads of money buying new faster kit. Also insomnia, needing frequent visits to the computer control centre, to check all units processing at maximum rate, welcome to the world of nerdy crunching, and blogging on strange websites with other compulsives. LoL
____________
|
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 882
ID: 105020
Credit: 1,318,826,036
RAC: 0
|
I think I've developed O C D, or obsessive Challenge disorder. Very worrying, causing excessive sweating over results, high temperatures on processing, and general desire to spend loads of money buying new faster kit. Also insomnia, needing frequent visits to the computer control centre, to check all units processing at maximum rate, welcome to the world of nerdy crunching, and blogging on strange websites with other compulsives. LoL
I worry I may catch that! Is it contagious?
I'm kidding, I've got that condition bad |
|
|
|
|
I worry I may catch that! Is it contagious?
I'm kidding, I've got that condition bad
We should care more about Coronavirus, many users are affected.
____________
My DC mathematical side :)
|
|
|
|
|
|
I found 1 user who downloaded tons of SR5. I don't know if it's good news.
E.g. 32-threads CPU: 25 valid, 707 in progress. :S
____________
My DC mathematical side :)
|
|
|
|
|
|
Here in China, the usual crowded streets are now empty; the buses often have one or two passengers only; and the markets close at 5PM!!
This is such madness I've never seen b4!(I wasn't alive during SARS)
People are staying at home and saying "O pigs I respect your ability to stay put forever!"
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
I think I've developed O C D, or obsessive Challenge disorder. Very worrying, causing excessive sweating over results, high temperatures on processing, and general desire to spend loads of money buying new faster kit. Also insomnia, needing frequent visits to the computer control centre, to check all units processing at maximum rate, welcome to the world of nerdy crunching, and blogging on strange websites with other compulsives. LoL
Really? If I had found a mega prime in this TdP, I'd be very satisfied.
People are staying at home and saying "O pigs I respect your ability to stay put forever!"
I think that one got lost in translation. https://en.wiktionary.org/wiki/pig#Etymology_1 |
|
|
|
|
|
And I found one. So glad to not miss the badge this year.
____________
|
|
|
|
|
[quote]I think I've developed O C D, or obsessive Challenge disorder. Very worrying, causing excessive sweating over results, high temperatures on processing, and general desire to spend loads of money buying new faster kit. Also insomnia, needing frequent visits to the computer control centre, to check all units processing at maximum rate, welcome to the world of nerdy crunching, and blogging on strange websites with other compulsives. LoL
Really? If I had found a mega prime in this TdP, I'd be very satisfied.
[quote]
Yes very happy, and perhaps it over encouraged my sense of humour.
____________
|
|
|
|
|
|
Homefarm wrote: ... If I had found a mega prime in this TdP, I'd be very satisfied. ...
If I find any prime that counts, I'll be very satisfied.
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment. |
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 882
ID: 105020
Credit: 1,318,826,036
RAC: 0
|
Here in China, the usual crowded streets are now empty; the buses often have one or two passengers only; and the markets close at 5PM!!
This is such madness I've never seen b4!(I wasn't alive during SARS)
People are staying at home and saying "O pigs I respect your ability to stay put forever!"
I hope you are safe and well
There have been difficulties in how China initially dealt with the virus by being mean through the police to Dr Lee who first reported it. I am sorry that Dr Lee has succumbed to the virus.
China now looks like they get the gravity of the situation. But I still get a sense that there is some priority on smoke and mirrors to deflect criticism.
I hope China may learn that their professionals, in their fields of expertise, are qualified enough to understand things better than the party |
|
|
|
|
|
Please stop be off topic.. and keep talk about "made in usa corona" at home in pentagon...))
here in norway no panic, no word about some "prevention" or scaring just in one "tabloid news"" as always ... just like in eu countries, only "panic news" do panic),, please by scientifist and pragmatic, chinese party do best, becouse they are simply best!
corona is great for just one evil- **** and debt economy ))
if you live in Gb better turn off sky news and other...))
keep calm and support Chinese products..)
cheers |
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 882
ID: 105020
Credit: 1,318,826,036
RAC: 0
|
keep calm and support Chinese products..)
cheers
I wish everyone a happy lunar new year!
Celebrations may be postponed, but it won't be forever.
Kung hei fat choy! |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
|
I added something new on everyone's private user page (home.php). Enjoy!
____________
My lucky number is 75898524288+1 |
|
|
|
|
I added something new on everyone's private user page (home.php). Enjoy!
You did and it's really cool! |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
I added something new on everyone's private user page (home.php). Enjoy!
And that's why PrimeGrid ranks first for useful add-ons to BOINC! :)
Hm, since you have the number of firsts I've done recently, you have the average size of work being done, and you have the sieve depth, can you calculate the "expected" time for me to find a prime in each project? As in "Expect 1 prime every: 80 days" or so? Can you calculate that overall, combining all projects being run? |
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 882
ID: 105020
Credit: 1,318,826,036
RAC: 0
|
I added something new on everyone's private user page (home.php). Enjoy!
And that's why PrimeGrid ranks first for useful add-ons to BOINC! :)
Hm, since you have the number of firsts I've done recently, you have the average size of work being done, and you have the sieve depth, can you calculate the "expected" time for me to find a prime in each project? As in "Expect 1 prime every: 80 days" or so? Can you calculate that overall, combining all projects being run?
That would be cool |
|
|
|
|
|
Thank you for that Micheal. Really nice quality of life improvement. |
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 882
ID: 105020
Credit: 1,318,826,036
RAC: 0
|
Thank you for that Micheal. Really nice quality of life improvement.
I second that! |
|
|
|
|
|
Greetings All
Well I finally got one (GFN-16) for 2020 TDP.
Also, thankyou Michael, that is some excellent information to watch.
Cheers |
|
|
|
|
|
For comps like this one, what % of first's should you be looking for?
I'm guessing its a bit subjective because more 1st's might come at a cost of throughput right?
Is there a "magic" number? More than 50% 1st's is ok I guess?
My GFN 17 Low is only producing 51.8% firsts..... I'm wondering if this is worthwhile?
____________
|
|
|
|
|
Here in China, the usual crowded streets are now empty; the buses often have one or two passengers only; and the markets close at 5PM!!
This is such madness I've never seen b4!(I wasn't alive during SARS)
People are staying at home and saying "O pigs I respect your ability to stay put forever!"
The same. I live in Jiangsu Province near Shanghai. I've been staying at home for 3 weeks without going out. The terrible virus caused this awkward situation. PrimeGrid and my pet would be my only companies. :( |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
Here in China, the usual crowded streets are now empty; the buses often have one or two passengers only; and the markets close at 5PM!!
This is such madness I've never seen b4!(I wasn't alive during SARS)
People are staying at home and saying "O pigs I respect your ability to stay put forever!"
The same. I live in Jiangsu Province near Shanghai. I've been staying at home for 3 weeks without going out. The terrible virus caused this awkward situation. PrimeGrid and my pet would be my only companies. :(
Please stay safe!
____________
My lucky number is 75898524288+1 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
I added something new on everyone's private user page (home.php). Enjoy!
And that's why PrimeGrid ranks first for useful add-ons to BOINC! :)
Hm, since you have the number of firsts I've done recently, you have the average size of work being done, and you have the sieve depth, can you calculate the "expected" time for me to find a prime in each project? As in "Expect 1 prime every: 80 days" or so? Can you calculate that overall, combining all projects being run?
I probably could, but I won't because it's just probabilities and I don't want to be repeating "Just because the odds are 1 in 15000, it doesn't mean something is broken because you've done 15001 tasks without finding anything".
Every day.
You know that will happen.
But it's very easy to estimate it on your own.
The odds of finding a prime are mostly determined by the size of the prime. The odds of finding a GFN16 or a PPSE are about 15,000 to 1. PPS-MEGA and 17_MEGA are about double that. You can extrapolate for even larger primes if you want.
____________
My lucky number is 75898524288+1 |
|
|
|
|
For comps like this one, what % of first's should you be looking for?
I'm guessing its a bit subjective because more 1st's might come at a cost of throughput right?
Is there a "magic" number? More than 50% 1st's is ok I guess?
My GFN 17 Low is only producing 51.8% firsts..... I'm wondering if this is worthwhile?
My GFN 17 Mega is at 47%. I'm still running it as I already have a P2020 badge, so working towards the Mega badge (which I won't get knowing my luck).
You could try GFN16 to see if your 1st's are higher if you are after more numbers on your P2020 badge. You'll also complete more tasks as it runs quicker, improving your chances. |
|
|
|
|
|
Thanks Mike for the great new function!!!!
Stay safe and always feel that you're going to find a prime quicker than me!
____________
SHSID Electronics Group
SHSIDElectronicsGroup@outlook.com
GFN-14: 50103906^16384+1
Proth "SoB": 44243*2^440969+1
|
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
For comps like this one, what % of first's should you be looking for?
I'm guessing its a bit subjective because more 1st's might come at a cost of throughput right?
Is there a "magic" number? More than 50% 1st's is ok I guess?
I've set up a whole system to determine what my best thread settings are. But a good rule-of-thumb seems to be that if your PPSE WUs are taking over 700 seconds, you might need more threads per WU. If under 400 seconds, maybe less threads.
Edit: Or maybe my computers are just slow. ;)
My GFN 17 Low is only producing 51.8% firsts..... I'm wondering if this is worthwhile?
What does the % of firsts on GFN matter? You can't give GFN work more than one GPU at a time, can you? |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
What does the % of firsts on GFN matter? You can't give GFN work more than one GPU at a time, can you?
It's the percentage of tasks that finished before your wingman, or the percentage of tasks where you would be the prime finder instead of the Double checker.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
New feature looks great! Thanks Mike.
____________
My lucky #: 60133106^131072+1 (GFN 17-mega) |
|
|
|
|
For comps like this one, what % of first's should you be looking for?
I'm guessing its a bit subjective because more 1st's might come at a cost of throughput right?
Is there a "magic" number? More than 50% 1st's is ok I guess?
I've set up a whole system to determine what my best thread settings are. But a good rule-of-thumb seems to be that if your PPSE WUs are taking over 700 seconds, you might need more threads per WU. If under 400 seconds, maybe less threads.
Edit: Or maybe my computers are just slow. ;)
My GFN 17 Low is only producing 51.8% firsts..... I'm wondering if this is worthwhile?
What does the % of firsts on GFN matter? You can't give GFN work more than one GPU at a time, can you?
I'm guessing that they may get better results moving down to GFN16 if faster cards are getting all the 1st's on a higher GFN. |
|
|
tngSend message
Joined: 29 Aug 10
Posts: 398
ID: 66603
Credit: 22,925,088,044
RAC: 2
|
For comps like this one, what % of first's should you be looking for?
I'm guessing its a bit subjective because more 1st's might come at a cost of throughput right?
Is there a "magic" number? More than 50% 1st's is ok I guess?
I've set up a whole system to determine what my best thread settings are. But a good rule-of-thumb seems to be that if your PPSE WUs are taking over 700 seconds, you might need more threads per WU. If under 400 seconds, maybe less threads.
Edit: Or maybe my computers are just slow. ;)
My GFN 17 Low is only producing 51.8% firsts..... I'm wondering if this is worthwhile?
What does the % of firsts on GFN matter? You can't give GFN work more than one GPU at a time, can you?
I'm guessing that they may get better results moving down to GFN16 if faster cards are getting all the 1st's on a higher GFN.
I figure anything over 50% is good. Remember that you will always have the ones where you got your task well after your wingman (aborts or timeouts by somebody else, or just luck), so that means you are enough better than average to offset that.
Haven't been running GFN-16, so don't know if that would be better. Lots of competition, but maybe fewer of the fast cards running it -- try it and see.
I've had 55.6% firsts on GFN-17 Mega, and 59.4% on GFN-18 -- less competition. 77.27% on GFN-19 -- chance of a prime showing up there is small, though.
____________
|
|
|
|
|
Thank you for that Micheal. Really nice quality of life improvement.
I second that!
I third it ;-)
____________
|
|
|
tngSend message
Joined: 29 Aug 10
Posts: 398
ID: 66603
Credit: 22,925,088,044
RAC: 2
|
For comps like this one, what % of first's should you be looking for?
I'm guessing its a bit subjective because more 1st's might come at a cost of throughput right?
Is there a "magic" number? More than 50% 1st's is ok I guess?
I've set up a whole system to determine what my best thread settings are. But a good rule-of-thumb seems to be that if your PPSE WUs are taking over 700 seconds, you might need more threads per WU. If under 400 seconds, maybe less threads.
Edit: Or maybe my computers are just slow. ;)
My GFN 17 Low is only producing 51.8% firsts..... I'm wondering if this is worthwhile?
What does the % of firsts on GFN matter? You can't give GFN work more than one GPU at a time, can you?
I'm guessing that they may get better results moving down to GFN16 if faster cards are getting all the 1st's on a higher GFN.
I figure anything over 50% is good. Remember that you will always have the ones where you got your task well after your wingman (aborts or timeouts by somebody else, or just luck), so that means you are enough better than average to offset that.
Haven't been running GFN-16, so don't know if that would be better. Lots of competition, but maybe fewer of the fast cards running it -- try it and see.
I've had 55.6% firsts on GFN-17 Mega, and 59.4% on GFN-18 -- less competition. 77.27% on GFN-19 -- chance of a prime showing up there is small, though.
Update: GFN-16 42.38%
GFN-17 Low 66.78%
____________
|
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 882
ID: 105020
Credit: 1,318,826,036
RAC: 0
|
Update: GFN-16 42.38%
GFN-17 Low 66.78%
I thought my stats of firsts were crap.
GFN-18 63%
This I have worked on and have a choice:
PPS-Mega 80%
|
|
|
Dave
Send message
Joined: 13 Feb 12
Posts: 2829
ID: 130544
Credit: 954,793,678
RAC: 0
|
|
GFN16 45%. Up from 44.44%. |
|
|
|
|
|
Nice display of personal stats.
Would it make sense for this table to be public info on account pages?
And a request for a global page showing a table of percentiles-vs-first percentage for each subproject, to provide context for the personal table.
The first column of each table lists first percentage (say in 5% increments, up to the maximum first percentage),
and the second column shows the percent of accounts (having a first percentage > 0 in that subproject)
whose first percentage exceeds the value in the first column.
Then 100% of users exceed 0% (so don't bother computing exceedance for 0%),
and (for example) if the maximum first percentage is 87%, then compute and display exceedance up to 85%.
It would be reasonable to update this page at a some interval, say daily or hourly, to reduce load on database server. |
|
|
TernVolunteer developer
Volunteer tester
Send message
Joined: 20 Sep 15
Posts: 26
ID: 421148
Credit: 306,661,290
RAC: 0
  
|
Nice display of personal stats.
Would it make sense for this table to be public info on account pages?
And a request for a global page showing a table of percentiles-vs-first percentage for each subproject, to provide context for the personal table.
Maybe I'm confused here... but wouldn't the overall average ALWAYS be 50%? Someone gets it in first, someone second. Obviously when you break it down by account, the person with the fastest hardware gets more firsts, but it's still going to total 50%. |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
My GFN 17 Low is only producing 51.8% firsts..... I'm wondering if this is worthwhile?
What does the % of firsts on GFN matter? You can't give GFN work more than one GPU at a time, can you?
I think we all misunderstood each other. Now that I've slept on it I'm looking at it a new way.
For GPUs, you can't do much to affect the First percentage statistic. So just ignore it! Look only at the Firsts-per-day statistic (as Michael mentioned earlier in this thread.) The higher your Firsts-per-day, the higher your chance of finding a prime first. A lower First percentage statistic doesn't change that at all, but does make it more likely for you to also show up on the doublecheckers list. ;)
This applies to CPUs as well. I've got one doing only 43% first (as estimated in my custom spreadsheet), but if I multithread it, the Firsts-per-day estimate goes down. So I just shouldn't worry about it. (Even though I do.) |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
Nice display of personal stats.
Would it make sense for this table to be public info on account pages?
And a request for a global page showing a table of percentiles-vs-first percentage for each subproject, to provide context for the personal table.
The first column of each table lists first percentage (say in 5% increments, up to the maximum first percentage),
and the second column shows the percent of accounts (having a first percentage > 0 in that subproject)
whose first percentage exceeds the value in the first column.
Then 100% of users exceed 0% (so don't bother computing exceedance for 0%),
and (for example) if the maximum first percentage is 87%, then compute and display exceedance up to 85%.
It would be reasonable to update this page at a some interval, say daily or hourly, to reduce load on database server.
Uh, no.
It's intentionally on only the private page for two reasons. The first is that I don't want to expose to everyone what you are doing. Your strategy is your strategy, and if you wish to keep that private, you should be able to keep it private. The second is performance. Public pages can be scraped by bots, and I don't want this query running continuously because of every wanna-be search engine that intentionally chooses to ignore our "no-robots" controls.
As for a big global table, that's beyond the scope of what I wanted to do here. This is a tool to help you tune your resources. It's not intended to give you insight into what everyone else is doing.
____________
My lucky number is 75898524288+1 |
|
|
|
|
Nice display of personal stats.
Would it make sense for this table to be public info on account pages?
And a request for a global page showing a table of percentiles-vs-first percentage for each subproject, to provide context for the personal table.
Maybe I'm confused here... but wouldn't the overall average ALWAYS be 50%? Someone gets it in first, someone second. Obviously when you break it down by account, the person with the fastest hardware gets more firsts, but it's still going to total 50%.
Percentile is usually an S-curve: a sloped section with a long near-horizontal tail at both ends. Are you saying that the inflection point of that curve is always at 50%? I disagree. In the extreme case, one user with a set of superfast CPUs can get up to 50% of all the tasks in a subproject (because of DC requirement) yet be the only person who has a first percentage > 5%. In that case it's useless for her to try to raise that first percentage to 10%.
EDIT:
Hmmm. In the extreme case cited, with 1000 users working on a subproject and rounding first-percentage to the nearest half percent, the second column would consist entirely of 0% because one person hogging all the firsts is just 0.1% of all users, and the other 999 users are all double-checkers. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 237,712,514
RAC: 0
|
For GPUs, you can't do much to affect the First percentage statistic.
That's news to me. :)
We just had a nice discussion on Discord yesterday about why someone running GFN-16 on a 1660TI was seeing 40% firsts while my two GPUs -- both slower than a 1660Ti -- were getting 50% (1060) and 70% (1660) firsts.
So, yes, there are things you can do that affect the first percentage, especially on short tasks like GFN-16s.
Setting your cache to 0 days is the obvious one. You should also set your project priority for PrimeGrid to 0 to eliminate BOINC's prefetch of tasks. On average, BOINC prefetches tasks 2.5 minutes before it is needed. When the takes only a couple of minutes to run the task, that prefetch time makes a big difference.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Michael Goetz wrote: ...
On average, BOINC prefetches tasks 2.5 minutes before it is needed. When the takes only a couple of minutes to run the task, that prefetch time makes a big difference.
Indeed.
My GTX 1060 3GB takes around 3min20sec per GFN-16 task.
A pre-fetch of 2,5 min knocks down the troughput and also the 1st percentage.
For a GFN-15 it'd be even worse.
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment. |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
