| Author |
Message |
|---|
 Michael Gutierrez Volunteer moderator
Project administrator
Project scientist
Send message
Joined: 21 Mar 17
Posts: 341
ID: 764476
Credit: 46,281,510
RAC: 0
                 
|
|
Welcome to the Geek Pride Day Challenge
The third challenge of the 2022 Series will be a 5-day challenge celebrating geeks, freaks, nerds, dorks, dweebs, and "weird" people of all kinds! The challenge will be offered on the GFN-19 subproject, beginning 25 May 18:00 UTC and ending 30 May 18:00 UTC.
May 25: ¡Feliz día del orgullo friki!
Geek Pride Day originated in 2006, when Spanish internet personality Señor Buebo created the Del Día Del Orgullo Friki website. It was originally a celebration of enthusiasts of traditionally-geeky franchises like Star Wars, Marvel Comics and so on. Today, though, when terms like "nerd" and "freak" are still sometimes used as insults, it's important to have a day to remember who we really are: people who are passionate about something! Whatever you love to geek out about, whether it's Minecraft, black holes, or prime numbers, you are part of one of the coolest, biggest, and weirdest communities in the universe!
  
To participate in the Challenge, please select only the GFN-19 subproject in your PrimeGrid preferences section.
Note: The deadline for some of these WUs is longer than five days, so make sure your computer is able to return the WUs within the challenge time-frame. Only tasks issued AFTER the start time and returned BEFORE the finish time will be counted.
Application Builds
IMPORTANT: Overclocking -- including factory overclocking -- on Nvidia GPUs is very strongly discouraged. Even if your GPU can run other tasks without difficulty, it may be unable to run GFN tasks when overclocked.
Supported platforms:
- Windows: Nvidia GPU (OpenCL): 32 bit, AMD/ATI GPU (OpenCL): 32 bit, CPU: 64 bit, 32 bit
- Linux: Nvidia GPU (OpenCL): 32 bit, 64 bit, AMD/ATI GPU (OpenCL): 32 bit, 64 bit, CPU: 64 bit, 32 bit
- Mac: Nvidia GPU (OpenCL): 64 bit, AMD/ATI GPU (OpenCL): 64 bit, CPU: 64 bit, 32 bit
A Cautionary Reminder
ATTENTION: The primality programs Genefer (for CPUs) and GeneferOCL (for GPUs) are computationally 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 CPU, and please see this post for tips on running GFN on your GPU successfully.
As with all number crunching, excessive heat can potentially cause permanent hardware failure. Please ensure your cooling system is sufficient.
GFN-19 WU's are currently averaging 40 hours on CPU (on one core) and 2 hours on GPU.
For a general idea of how your GPU stacks up, you can have a look at the fastest GPUs list.
If your CPU is highly overclocked, please consider "stress testing" it. Overclocking your GPU is not recommended at all for GeneferOCL. Sieving is an excellent alternative for computers that are not able to run Genefer. :)
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 May 25th 18:00 UTC and received BEFORE May 30th 18:00 UTC will be considered for credit.
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 alows them to be recycled immediately; thus a much faster "clean up" to the end of a 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. :)
More information on Generalized Fermat Numbers and the Genefer program
Best of Luck to everyone!
____________
Time flies like an arrow. Fruit flies like a banana.
|
|
|
|
|
|
Any new GFN-19 would come in on Top 60 of largest known primes.
Currently known and verified GFN-19 can be seen on
https://primes.utm.edu/primes/search.php?Description=%25%5E524288%2B1%24&Style=HTML
and
https://primes.utm.edu/top20/page.php?id=12.
Unusually many of them were found in 2018-2019. It will be two years since the last find, during the challenge.
/JeppeSN |
|
|
Yves GallotVolunteer developer
Project scientist
Send message
Joined: 19 Aug 12
Posts: 800
ID: 164101
Credit: 305,637,084
RAC: 5,220
|
|
Here is some information on multi-threading.
The allocated data size is 10 MB.
On processors with two memory channels, the throughput is generally optimal if the data size of all tasks is lower than the L3 cache size. For example, i7-11700K: L3 = 16 MB => 1 task; i7-12700K: L3 = 25 MB => 1 or 2 tasks; Ryzen 5950X: L3 = 2 x 32 MB => 6 tasks or less.
If the L3 cache size is below 10 MB, hyper-threading may increase the speed of genefer because two threads per core hide the memory latency.
Processors with four memory channels can be more efficient if the data size is greater than the L3 cache size.
On Ryzen processors with two CCX, it is recommanded that the affinity of genefer threads is set (see http://www.primegrid.com/forum_thread.php?id=9905). |
|
|
|
|
|
An early GFN-19: 4896418^524288 + 1. /JeppeSN |
|
|
|
|
An early GFN-19: 4896418^524288 + 1. /JeppeSN
NICE! |
|
|
|
|
|
So we did not quite reach the two-years mark before another one was found. /JeppeSN |
|
|
|
|
|
The last one I found: 2020-05-29 07:52:25 UTC
Quite close to two years.
____________
DeleteNull |
|
|
|
|
Here is some information on multi-threading.
Is this about multi-threading or just about the number of tasks?
____________
Greetings, Jens
147433824^131072+1 |
|
|
Yves GallotVolunteer developer
Project scientist
Send message
Joined: 19 Aug 12
Posts: 800
ID: 164101
Credit: 305,637,084
RAC: 5,220
|
Here is some information on multi-threading.
Is this about multi-threading or just about the number of tasks?
??? Both: number of tasks x number of threads = number of cores. |
|
|
|
|
Here is some information on multi-threading.
Is this about multi-threading or just about the number of tasks?
??? Both: number of tasks x number of threads = number of cores.
I thought only GFN-21 could be multi-threaded.
____________
Greetings, Jens
147433824^131072+1 |
|
|
|
|
|
I believe it is a very recent addition. You control it through the MT control on the website, just like LLR. |
|
|
|
|
|
Seeing way to many Aborted Tasks as it seems to be the norm before a Challenge Series starts now. |
|
|
|
|
I believe it is a very recent addition. You control it through the MT control on the website, just like LLR.
I see.
Thx.
____________
Greetings, Jens
147433824^131072+1 |
|
|
|
|
|
Do I need to set max cpu threads to 1 for this challenge or is no limit better? |
|
|
ReggieVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 10 May 14
Posts: 230
ID: 311759
Credit: 206,907,978
RAC: 46,995
                 
|
|
Last few posts are getting off topic folks... Consider this a warning before I start removing things. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 390,667,618
RAC: 107,419
                         
|
Last few posts are getting off topic folks... Consider this a warning before I start removing things.
This.
Lots of posts have been removed for being off topic.
When you post anything here, there's a bunch of rules prominently displayed just to the left of where you type the message. Besides being off topic, many of the posts also ran afoul of these rules, or came really close:
- No messages intended to annoy or antagonize other people.
- No messages that are deliberately hostile or insulting.
- No abusive comments involving race, religion, nationality, gender, class or sexuality.
- No discussions about religion or politics.
Play nice... or play somewhere else, please.
____________
My lucky number is 75898524288+1
|
|
|
Michael GutierrezVolunteer moderator
Project administrator
Project scientist
Send message
Joined: 21 Mar 17
Posts: 341
ID: 764476
Credit: 46,281,510
RAC: 0
|
|
Day 1 is done, here are the stats!
Challenge: Geek Pride Day
App: 27 (GFN-19)
(As of 2022-05-26 21:57:30 UTC)
50502 tasks have been sent out. [CPU/GPU/anonymous_platform: 14879 (29%) / 35623 (71%) / 0 (0%)]
Of those tasks that have been sent out:
6254 (12%) were aborted. [1850 (4%) / 4404 (9%) / 0 (0%)]
2106 (4%) came back with some kind of an error. [494 (1%) / 1612 (3%) / 0 (0%)]
21603 (43%) have returned a successful result. [4367 (9%) / 17236 (34%) / 0 (0%)]
20539 (41%) are still in progress. [8168 (16%) / 12371 (24%) / 0 (0%)]
Of the tasks that have been returned successfully:
8986 (42%) are pending validation. [1846 (9%) / 7140 (33%) / 0 (0%)]
12607 (58%) have been successfully validated. [2513 (12%) / 10094 (47%) / 0 (0%)]
8 (0%) were invalid. [7 (0%) / 1 (0%) / 0 (0%)]
2 (0%) are inconclusive. [1 (0%) / 1 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is b=5083164. The leading edge was at b=4950146 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 2.69% as much as it had prior to the challenge!
____________
Time flies like an arrow. Fruit flies like a banana. |
|
|
Michael GutierrezVolunteer moderator
Project administrator
Project scientist
Send message
Joined: 21 Mar 17
Posts: 341
ID: 764476
Credit: 46,281,510
RAC: 0
|
|
Day 2 stats!
Challenge: Geek Pride Day
App: 27 (GFN-19)
(As of 2022-05-27 22:17:38 UTC)
80591 tasks have been sent out. [CPU/GPU/anonymous_platform: 23213 (29%) / 57378 (71%) / 0 (0%)]
Of those tasks that have been sent out:
9729 (12%) were aborted. [3553 (4%) / 6176 (8%) / 0 (0%)]
4341 (5%) came back with some kind of an error. [496 (1%) / 3845 (5%) / 0 (0%)]
45120 (56%) have returned a successful result. [9852 (12%) / 35268 (44%) / 0 (0%)]
21401 (27%) are still in progress. [9312 (12%) / 12089 (15%) / 0 (0%)]
Of the tasks that have been returned successfully:
12925 (29%) are pending validation. [2820 (6%) / 10105 (22%) / 0 (0%)]
32178 (71%) have been successfully validated. [7021 (16%) / 25157 (56%) / 0 (0%)]
15 (0%) were invalid. [11 (0%) / 4 (0%) / 0 (0%)]
2 (0%) are inconclusive. [0 (0%) / 2 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is b=5156284. The leading edge was at b=4950146 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 4.16% as much as it had prior to the challenge!
____________
Time flies like an arrow. Fruit flies like a banana. |
|
|
|
|
|
Is 12% high for aborted Tasks? |
|
|
|
|
Is 12% high for aborted Tasks?
Doesn't really matter. Aborted tasks are probably from people who accidentally downloaded too many tasks and aborted them so they have a better chance of returning a result 1st and getting credit for finding a prime instead of being the DC'er.
Doesn't change anything in the challenge.
Aborted tasks are better than abandoned tasks. IE: Tasks that are removed by means other than the BOINC client whether it's deleting them from the host directly or issuing a project reset on the BOINC client.
Either way, it doesn't effect the scoring at all here.
Only negative effect is the person who downloads the resends will almost definitely be a DC'er. |
|
|
|
|
|
It does stink getting a lot of the recycled tasks if you really like being the prime finder. It is kind of why people look for primes to begin with.
But not much anyone can do about it.
Everyone has the same odds of getting said tasks so the best thing to do is not look at why you’re not the first returned.
Just look at what percent first you are and leave it at that.
Wish I could join in on this one.
Good luck finding some primes
See you next challenge. |
|
|
Michael GutierrezVolunteer moderator
Project administrator
Project scientist
Send message
Joined: 21 Mar 17
Posts: 341
ID: 764476
Credit: 46,281,510
RAC: 0
|
|
Less than 48 hours remain! Daily stats:
Challenge: Geek Pride Day
App: 27 (GFN-19)
(As of 2022-05-28 19:01:40 UTC)
107522 tasks have been sent out. [CPU/GPU/anonymous_platform: 28989 (27%) / 78533 (73%) / 0 (0%)]
Of those tasks that have been sent out:
11341 (11%) were aborted. [3767 (4%) / 7574 (7%) / 0 (0%)]
7025 (7%) came back with some kind of an error. [505 (0%) / 6520 (6%) / 0 (0%)]
66959 (62%) have returned a successful result. [14759 (14%) / 52200 (49%) / 0 (0%)]
22197 (21%) are still in progress. [9958 (9%) / 12239 (11%) / 0 (0%)]
Of the tasks that have been returned successfully:
14786 (22%) are pending validation. [3156 (5%) / 11630 (17%) / 0 (0%)]
52150 (78%) have been successfully validated. [11590 (17%) / 40560 (61%) / 0 (0%)]
21 (0%) were invalid. [13 (0%) / 8 (0%) / 0 (0%)]
2 (0%) are inconclusive. [0 (0%) / 2 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is b=5229110. The leading edge was at b=4950146 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 5.64% as much as it had prior to the challenge!
____________
Time flies like an arrow. Fruit flies like a banana. |
|
|
|
|
Is 12% high for aborted Tasks?
Doesn't really matter. Aborted tasks are probably from people who accidentally downloaded too many tasks and aborted them so they have a better chance of returning a result 1st and getting credit for finding a prime instead of being the DC'er.
Doesn't change anything in the challenge.
Aborted tasks are better than abandoned tasks. IE: Tasks that are removed by means other than the BOINC client whether it's deleting them from the host directly or issuing a project reset on the BOINC client.
Either way, it doesn't effect the scoring at all here.
Only negative effect is the person who downloads the resends will almost definitely be a DC'er.
Well my review is they are members that know what they are doing and doing before each Challenge.
You do not download 900 tasks and then aborted them and do the same the next day.
Most aborted tasks are Second run tasks and will not be 1st that is why I think know long time members do this.
I just do not care to complete tasks that have been aborted a day ago knowing that the tasks was completed already.
"Doesn't change anything in the challenge." But it is not all about the challenge but about finding the Prime.....
But I also see your Point. |
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2379
ID: 1178
Credit: 17,806,088,607
RAC: 5,368,922
                                              
|
Is 12% high for aborted Tasks?
Doesn't really matter. Aborted tasks are probably from people who accidentally downloaded too many tasks and aborted them so they have a better chance of returning a result 1st and getting credit for finding a prime instead of being the DC'er.
Doesn't change anything in the challenge.
Aborted tasks are better than abandoned tasks. IE: Tasks that are removed by means other than the BOINC client whether it's deleting them from the host directly or issuing a project reset on the BOINC client.
Either way, it doesn't effect the scoring at all here.
Only negative effect is the person who downloads the resends will almost definitely be a DC'er.
Well my review is they are members that know what they are doing and doing before each Challenge.
You do not download 900 tasks and then aborted them and do the same the next day.
Most aborted tasks are Second run tasks and will not be 1st that is why I think know long time members do this.
I just do not care to complete tasks that have been aborted a day ago knowing that the tasks was completed already.
"Doesn't change anything in the challenge." But it is not all about the challenge but about finding the Prime.....
But I also see your Point.
Keep in mind that badly updated drivers can cause lots of errors on GPU challenges. I have generally avoided that on my Windows machines, but it happened on one Ubuntu Linux box this time around that I did not notice for a bit. Can run through hundreds of erroring work units fairly quickly.
|
|
|
|
|
|
Maybe they had large caches for the pentathlon and forgot to set it to zero for the PG challenge
Either way everyone has the same odds of getting those tasks if you’re set to zero cache. |
|
|
|
|
Is 12% high for aborted Tasks?
Doesn't really matter. Aborted tasks are probably from people who accidentally downloaded too many tasks and aborted them so they have a better chance of returning a result 1st and getting credit for finding a prime instead of being the DC'er.
Doesn't change anything in the challenge.
Aborted tasks are better than abandoned tasks. IE: Tasks that are removed by means other than the BOINC client whether it's deleting them from the host directly or issuing a project reset on the BOINC client.
Either way, it doesn't effect the scoring at all here.
Only negative effect is the person who downloads the resends will almost definitely be a DC'er.
Well my review is they are members that know what they are doing and doing before each Challenge.
You do not download 900 tasks and then aborted them and do the same the next day.
Most aborted tasks are Second run tasks and will not be 1st that is why I think know long time members do this.
I just do not care to complete tasks that have been aborted a day ago knowing that the tasks was completed already.
"Doesn't change anything in the challenge." But it is not all about the challenge but about finding the Prime.....
But I also see your Point.
Well if finding a prime is more important than the Challenge Series then perhaps running a sub project that isn't in the Challenge Series will net more 1sts.
I get what you're saying though. And only the admins would be able to look into that |
|
|
|
|
Is 12% high for aborted Tasks?
Doesn't really matter. Aborted tasks are probably from people who accidentally downloaded too many tasks and aborted them so they have a better chance of returning a result 1st and getting credit for finding a prime instead of being the DC'er.
Doesn't change anything in the challenge.
Aborted tasks are better than abandoned tasks. IE: Tasks that are removed by means other than the BOINC client whether it's deleting them from the host directly or issuing a project reset on the BOINC client.
Either way, it doesn't effect the scoring at all here.
Only negative effect is the person who downloads the resends will almost definitely be a DC'er.
Well my review is they are members that know what they are doing and doing before each Challenge.
You do not download 900 tasks and then aborted them and do the same the next day.
Most aborted tasks are Second run tasks and will not be 1st that is why I think know long time members do this.
I just do not care to complete tasks that have been aborted a day ago knowing that the tasks was completed already.
"Doesn't change anything in the challenge." But it is not all about the challenge but about finding the Prime.....
But I also see your Point.
Keep in mind that badly updated drivers can cause lots of errors on GPU challenges. I have generally avoided that on my Windows machines, but it happened on one Ubuntu Linux box this time around that I did not notice for a bit. Can run through hundreds of erroring work units fairly quickly.
Yes I do see hundreds of errors and it seems that the members is not managing their own hardware. But that has been a problem here as well as all other BOINC Projects. |
|
|
|
|
|
To get slightly more points and a few ranks, participants get a lot of tasks, then abort the less advantageous.
This could be avoided by an amount of point per task fixed in advance for the whole challenge. |
|
|
|
|
To get slightly more points and a few ranks, participants get a lot of tasks, then abort the less advantageous.
Why do you think some tasks are more advantageous? The typical (and intended) situation is that a task that gives slightly more points also takes slightly longer to run. The points awarded should reflect the "work" done.
This could be avoided by an amount of point per task fixed in advance for the whole challenge.
The tasks get larger and larger. Tasks sent out in the beginning of a challenge are generally a bit "easier" (because b is smaller) than tasks sent out in the end. If you gave them the same score, a rational person who wanted to maximize his challenge score (and who did not care about being 1st or 2nd!) would bunker the easier tasks from the beginning. So the opposite consequence of what you wanted to achieve.
/JeppeSN |
|
|
|
|
|
You're right ! No solution. |
|
|
|
|
|
It has been said many times before, but let me repeat.
If some user X takes and holds a task (or a lot of tasks) without correctly completing it, this is not to the advantage of that user in any way.
So how does it affect other users? In one sense, it does not affect them because their computers will crunch just as efficiently, with no interruptions, as if the user X had not existed. So for challenge points, it is entirely irrelevant.
But what about the implications on other users wrt. being first or not? Here, the behavior of X can give an advantage to a random (slow) host (excluding hosts belonging to X), and give a disadvantage to a random (fast) host.
When the work unit is new, typically two tasks are generated with names ending in _0 and _1. Often the two tasks are sent out at about the same instance of time (it is random if _1 or _0 goes out first). Let us say the careless user X mentioned gets one of them, and some random user A gets the other one. Now A has an advantage in becoming 1st, because X is not going to complete his tasks correctly, so A can easily get first. Even if A has a slow computer, they can get first.
When poor user X no longer holds the task (it is aborted, or has errored, or has timed out), BOINC software on the server realizes the need for a third task, with a name ending in _2. This tasks is sent to a random user, call them B. So B gets a disadvantage, and will often become 2nd. Even if B has a very fast computer, they can get second.
So you can become A or B, with equal probability. If all your computers are slow, it is good news; when you are A, you can still become 1st with your slow hardware. If all your computers are very fast, it is bad news; when you are B, you can lose even if your computer is faster.
Overall, the consequence of user X's carelessness is that slow computers get a random advantage over fast computers.
It is often praised as something positive that a very slow computer can still be lucky and win (become A). But if you are the person with all the fastest machines, sometimes a 1st is "stolen" from you in this way (you are B).
/JeppeSN |
|
|
|
|
|
But, if you're C and get an aborted task ending by _2 or _3 ?
An what should happen to X ?
There is no first if the task was aborted ? It happens more often than one would believe.
Definitively, the algorithm and criteria to be first should be detailed.
First to return ? Minimum run time ? Date received ?
To be first or not to be first ... |
|
|
|
|
|
The definition of "first" is the one to first return a correct result to the server. Sometimes, the person who gets first had an advantage, sometimes they even had a disadvantage.
When it goes to tasks _3, _4, etc., it is the same as with _2. No matter how large your "serial" number "_n" in the name is, you cannot know for sure if you are really an A or a B. For example, with wuid=798450888, the person who got the _2 task (third row) had a safe win. They were an A there.
(If they had aborted because the name of their task ended in _2, they would have lost a sure win, and then the _3 would have become an "A" task instead.)
/JeppeSN
|
|
|
|
|
|
Thank you for your explanations, JeppeSN. It's a lot more clearer. |
|
|
mikey
Send message
Joined: 17 Mar 09
Posts: 1645
ID: 37043
Credit: 733,135,347
RAC: 134,296
      
|
The definition of "first" is the one to first return a correct result to the server. Sometimes, the person who gets first had an advantage, sometimes they even had a disadvantage.
When it goes to tasks _3, _4, etc., it is the same as with _2. No matter how large your "serial" number "_n" in the name is, you cannot know for sure if you are really an A or a B. For example, with wuid=798450888, the person who got the _2 task (third row) had a safe win. They were an A there.
(If they had aborted because the name of their task ended in _2, they would have lost a sure win, and then the _3 would have become an "A" task instead.)
/JeppeSN
You keep calling it "a win" if you return it first, I don't think of it that way. I think of "a win" as me successfully completing the task whether I am first or last or somewhere in between. I have slow computers and faster computers but can't compete with the people pushing the envelope and always buying the latest and greatest hardware and maybe that's why I think of "a win" the way I do.
Now have I been first before of course and everyone will at some point as you pointed out even thru dumb luck, but it takes two results to validate "a win" and without the 2nd result it's not "a win" but instead just a result. |
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 2209
ID: 105020
Credit: 8,124,218,773
RAC: 1,940,528
                        
|
|
I consider it a win if I am a part of a find at all.
This is probably coming from a place where I cannot (or wish not) be connected to the internet.
It would be better to be 2nd than nothing?
I may have repeated something - I skimmed over it.
Or I have a lunaticit approach - self evident from the new word - with all things it is an amalgamation of things being ideas - usually a dialectic. |
|
|
|
|
|
#29 is fine, #125 too, in the metaverse. |
|
|
Michael GutierrezVolunteer moderator
Project administrator
Project scientist
Send message
Joined: 21 Mar 17
Posts: 341
ID: 764476
Credit: 46,281,510
RAC: 0
|
|
1 day remains...
Challenge: Geek Pride Day
App: 27 (GFN-19)
(As of 2022-05-29 18:07:45 UTC)
133052 tasks have been sent out. [CPU/GPU/anonymous_platform: 35260 (27%) / 97792 (73%) / 0 (0%)]
Of those tasks that have been sent out:
12144 (9%) were aborted. [4423 (3%) / 7721 (6%) / 0 (0%)]
7574 (6%) came back with some kind of an error. [506 (0%) / 7068 (5%) / 0 (0%)]
90997 (68%) have returned a successful result. [20337 (15%) / 70660 (53%) / 0 (0%)]
22337 (17%) are still in progress. [9994 (8%) / 12343 (9%) / 0 (0%)]
Of the tasks that have been returned successfully:
15807 (17%) are pending validation. [3416 (4%) / 12391 (14%) / 0 (0%)]
75165 (83%) have been successfully validated. [16906 (19%) / 58259 (64%) / 0 (0%)]
25 (0%) were invalid. [15 (0%) / 10 (0%) / 0 (0%)]
0 (0%) are inconclusive. [0 (0%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is b=5306862. The leading edge was at b=4950146 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 7.21% as much as it had prior to the challenge!
____________
Time flies like an arrow. Fruit flies like a banana. |
|
|
|
|
When the work unit is new, typically two tasks are generated with names ending in _0 and _1. Often the two tasks are sent out at about the same instance of time (it is random if _1 or _0 goes out first). Let us say the careless user X mentioned gets one of them, and some random user A gets the other one. Now A has an advantage in becoming 1st, because X is not going to complete his tasks correctly, so A can easily get first. Even if A has a slow computer, they can get first.
/JeppeSN
This here should clear it up then.
If I'm not mistaken bcavnaugh is assuming that the task ending in _1 was the second task of that work unit to be sent. If you are saying that it's random that _1 or _0 is sent first then no user will be able to have an advantage by aborting all tasks ending in _1.
The advantage being returning the task first because they received the task first. When being the first to return a task is the goal.
However, for the purpose of winning the competition it doesn't matter as long as the task was downloaded at the start of the challenge or somewhere in between the start and end. Then returned before the end of the challenge. Additionally, that returned task must be a valid task. |
|
|
|
|
When the work unit is new, typically two tasks are generated with names ending in _0 and _1. Often the two tasks are sent out at about the same instance of time (it is random if _1 or _0 goes out first). Let us say the careless user X mentioned gets one of them, and some random user A gets the other one. Now A has an advantage in becoming 1st, because X is not going to complete his tasks correctly, so A can easily get first. Even if A has a slow computer, they can get first.
/JeppeSN
This here should clear it up then.
If I'm not mistaken bcavnaugh is assuming that the task ending in _1 was the second task of that work unit to be sent. If you are saying that it's random that _1 or _0 is sent first then no user will be able to have an advantage by aborting all tasks ending in _1.
The advantage being returning the task first because they received the task first. When being the first to return a task is the goal.
However, for the purpose of winning the competition it doesn't matter as long as the task was downloaded at the start of the challenge or somewhere in between the start and end. Then returned before the end of the challenge. Additionally, that returned task must be a valid task.
@Skillz: You're the only first!
As first, do you care of being first to return tasks because your probability to find a prime is optimal ?
What's is your overall first percentage ? Has it changed since the beginning of the challenge ?
Note that in french "premier" means first but also prime! |
|
|
|
|
@Skillz: You're the only first!
As first, do you care of being first to return tasks because your probability to find a prime is optimal ?
What's is your overall first percentage ? Has it changed since the beginning of the challenge ?
Note that in french "premier" means first but also prime!
I don't know what my overall is. I've got a bunch of 3090s in the race and all of them are 90%+ first rate. I've also got some really slow 1080 and 1070 cards that are around 50%. |
|
|
|
|
@Skillz: You're the only first!
As first, do you care of being first to return tasks because your probability to find a prime is optimal ?
What's is your overall first percentage ? Has it changed since the beginning of the challenge ?
Note that in french "premier" means first but also prime!
I don't know what my overall is. I've got a bunch of 3090s in the race and all of them are 90%+ first rate. I've also got some really slow 1080 and 1070 cards that are around 50%.
Quite impressive! Glad to doublecheck/second a few of your firsts.
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13951
ID: 53948
Credit: 390,667,618
RAC: 107,419
|
You're right ! No solution.
It’s not so much that there’s no solution as there is no problem.
For GFN the run times are very predictable and, within each subproject, the credit granted is directly proportional to the time it takes to run the task. No matter which GFN19 task you get, you will receive exactly the same credit/hour. It’s designed specifically to work that way. I should know; I am the person who designed it.
____________
My lucky number is 75898524288+1 |
|
|
|
|
This here should clear it up then.
If I'm not mistaken bcavnaugh is assuming that the task ending in _1 was the second task of that work unit to be sent. If you are saying that it's random that _1 or _0 is sent first then no user will be able to have an advantage by aborting all tasks ending in _1.
You can just look at the WU (work unit) I linked in my previous post. The _0 task (top row) was sent out at "25 May 2022 | 18:16:32 UTC". The _1 task (second row) was sent "25 May 2022 | 18:13:36 UTC". So the _1 had an advantage of 2 minutes 56 seconds over the _0. /JeppeSN
|
|
|
|
|
To get slightly more points and a few ranks, participants get a lot of tasks, then abort the less advantageous.
I completely agree. The sentence above is false. It is pointless to select among tasks. The design is fair.
You brightly demonstrated that there is no advantage to abort tasks.
|
|
|
MonkeydeeVolunteer tester
Send message
Joined: 8 Dec 13
Posts: 532
ID: 284516
Credit: 1,413,716,796
RAC: 651,903
               
|
|
Everyone notices when the 1st rate goes down. So they investigate and see that they're DCing work from someone who aborted a pile of tasks.
Most don't notice when their 1st rate goes up. So they don't see that they were the 1st while the original DCer aborted a pile of tasks.
In theory it should even out over time between the two situations.
A challenge can actually improve things as there are a wider variety of people running the same type of task so you're more likely to be doing tasks with reliable hosts of all kinds of different speeds.
The situation I'm not a huge fan of is when one person throws as insane amount of horsepower at a project that very few other people are running. That causes a situation of one person "pulling out ahead" until the server says "no tasks for you" and everyone else is DCing that person. However, this situation is completely nullified in a challenge due to the insanely wide variety of different people running the same project.
____________
My Primes
Badge Score: 4*2 + 6*2 + 7*4 + 8*8 + 11*3 + 12*1 = 157
 |
|
|
|
|
|
You mean like this user https://www.primegrid.com/results.php?hostid=932196&offset=0&show_names=0&state=7&appid=27
900+ Errors
In all over 3500 errors https://www.primegrid.com/results.php?hostid=932196&offset=0&show_names=0&state=7&appid=0
Thank you Gridcoin |
|
|
|
|
Everyone notices when the 1st rate goes down. So they investigate and see that they're DCing work from someone who aborted a pile of tasks.
Most don't notice when their 1st rate goes up. So they don't see that they were the 1st while the original DCer aborted a pile of tasks.
In theory it should even out over time between the two situations.
A challenge can actually improve things as there are a wider variety of people running the same type of task so you're more likely to be doing tasks with reliable hosts of all kinds of different speeds.
The situation I'm not a huge fan of is when one person throws as insane amount of horsepower at a project that very few other people are running. That causes a situation of one person "pulling out ahead" until the server says "no tasks for you" and everyone else is DCing that person. However, this situation is completely nullified in a challenge due to the insanely wide variety of different people running the same project.
During the TDP, I noticed my 1st rate decline when tng put everything onto the same sub project lol.
I was the DC for just about all my tasks.
I just buy my own hardware and crunch. That's it. BOINC is a bit of a rich man's hobby if you want to live on the bleeding edge of technology and stay at the front of the pack. That can now mean buying the latest & greatest but also mean being able to afford the power bill :(
I don't get my hopes up of finding a prime, that way I'm never disappointed.
____________
Слава Україні! |
|
|
Dave
Send message
Joined: 13 Feb 12
Posts: 3168
ID: 130544
Credit: 2,224,464,176
RAC: 228,967
           
|
To get slightly more points and a few ranks, participants get a lot of tasks, then abort the less advantageous.
May or may not work. It's a lottery. Cherry-picking smaller tasks may mean you can fit 1 more in right before the end. Or be unlucky & just miss that opportunity. Best to just leave it & let it flow naturally. |
|
|
|
|
|
My rate dropped from 88-92 down to 60 overnight do to so many Error and aborted by user on the 28th and 29th even hundreds of Errors from good known members. |
|
|
|
|
My rate dropped from 88-92 down to 60 overnight do to so many Error and aborted by user on the 28th and 29th even hundreds of Errors from good known members.
Of all the tasks you did that were errors or abandoned, were any of them prime? |
|
|
|
|
My rate dropped from 88-92 down to 60 overnight do to so many Error and aborted by user on the 28th and 29th even hundreds of Errors from good known members.
Of all the tasks you did that were errors or abandoned, were any of them prime?
From what I see we did not find any Primes this year, but we will have to wait and see.
There are to many to review so I would not know myself. |
|
|
Michael GutierrezVolunteer moderator
Project administrator
Project scientist
Send message
Joined: 21 Mar 17
Posts: 341
ID: 764476
Credit: 46,281,510
RAC: 0
|
|
We're all done! Here are the final stats. Cleanup starting now!
Challenge: Geek Pride Day
App: 27 (GFN-19)
(As of 2022-05-30 19:39:15 UTC)
158005 tasks have been sent out. [CPU/GPU/anonymous_platform: 40338 (26%) / 117667 (74%) / 0 (0%)]
Of those tasks that have been sent out:
16459 (10%) were aborted. [7613 (5%) / 8846 (6%) / 0 (0%)]
8431 (5%) came back with some kind of an error. [510 (0%) / 7921 (5%) / 0 (0%)]
117458 (74%) have returned a successful result. [26260 (17%) / 91198 (58%) / 0 (0%)]
14138 (9%) are still in progress. [5455 (3%) / 8682 (5%) / 0 (0%)]
Of the tasks that have been returned successfully:
11812 (10%) are pending validation. [2630 (2%) / 9182 (8%) / 0 (0%)]
105616 (90%) have been successfully validated. [23613 (20%) / 82003 (70%) / 0 (0%)]
30 (0%) were invalid. [17 (0%) / 13 (0%) / 0 (0%)]
0 (0%) are inconclusive. [0 (0%) / 0 (0%) / 0 (0%)]
The current leading edge (i.e., latest work unit for which work has actually been sent out to a host) is b=5372518. The leading edge was at b=4950146 at the beginning of the challenge. Since the challenge started, the leading edge has advanced 8.53% as much as it had prior to the challenge!
____________
Time flies like an arrow. Fruit flies like a banana. |
|
|
|
|
Is 12% high for aborted Tasks?
Doesn't really matter. Aborted tasks are probably from people who accidentally downloaded too many tasks and aborted them so they have a better chance of returning a result 1st and getting credit for finding a prime instead of being the DC'er.
Doesn't change anything in the challenge.
Aborted tasks are better than abandoned tasks. IE: Tasks that are removed by means other than the BOINC client whether it's deleting them from the host directly or issuing a project reset on the BOINC client.
Either way, it doesn't effect the scoring at all here.
Only negative effect is the person who downloads the resends will almost definitely be a DC'er.
Well my review is they are members that know what they are doing and doing before each Challenge.
You do not download 900 tasks and then aborted them and do the same the next day.
Most aborted tasks are Second run tasks and will not be 1st that is why I think know long time members do this.
I just do not care to complete tasks that have been aborted a day ago knowing that the tasks was completed already.
"Doesn't change anything in the challenge." But it is not all about the challenge but about finding the Prime.....
But I also see your Point.
Boinc is terrible at judging how many tasks to download with regards to multithreaded apps. Add into the fact that it also is terrible at judging task completion times for gpu and cpu tasks from the same project, and you end up with people like me who request 1 day worth of work and end up with 2 weeks worth.
It's nothing malicious. |
|
|
Nick
Send message
Joined: 11 Jul 11
Posts: 2209
ID: 105020
Credit: 8,124,218,773
RAC: 1,940,528
|
Is 12% high for aborted Tasks?
Doesn't really matter. Aborted tasks are probably from people who accidentally downloaded too many tasks and aborted them so they have a better chance of returning a result 1st and getting credit for finding a prime instead of being the DC'er.
Doesn't change anything in the challenge.
Aborted tasks are better than abandoned tasks. IE: Tasks that are removed by means other than the BOINC client whether it's deleting them from the host directly or issuing a project reset on the BOINC client.
Either way, it doesn't effect the scoring at all here.
Only negative effect is the person who downloads the resends will almost definitely be a DC'er.
Well my review is they are members that know what they are doing and doing before each Challenge.
You do not download 900 tasks and then aborted them and do the same the next day.
Most aborted tasks are Second run tasks and will not be 1st that is why I think know long time members do this.
I just do not care to complete tasks that have been aborted a day ago knowing that the tasks was completed already.
"Doesn't change anything in the challenge." But it is not all about the challenge but about finding the Prime.....
But I also see your Point.
Boinc is terrible at judging how many tasks to download with regards to multithreaded apps. Add into the fact that it also is terrible at judging task completion times for gpu and cpu tasks from the same project, and you end up with people like me who request 1 day worth of work and end up with 2 weeks worth.
It's nothing malicious.
Or you request a day and get a minute.
It is usually the other way round.
And it is not consistent.
One day 0.3 days equals 2 days.
Then next time, 0.6 days equals 0.2 days.
This may depend on what is running and how many of them.
This is how it is.
Edit: It hasn't been that wildly different. But it has been a lot different and not closely related to what is actually going on. |
|
|
|
|
|
A special acknowledgement to Yves Gallot and the other team scientists for the recent enhancements introduced in genefer, although the official announcement will come after post-challenge feedback ?
Merci Yves!
|
|
|
|
|
A special acknowledgement to Yves Gallot and the other team scientists for the recent enhancements introduced in genefer, although the official announcement will come after post-challenge feedback ?
Merci Yves!
Oui oui!
CPU MT GFN19 really helped me in the challenge-merci beaucoup!!
____________
My lucky number is 6219*2^3374198+1
 |
|
|
Yves GallotVolunteer developer
Project scientist
Send message
Joined: 19 Aug 12
Posts: 800
ID: 164101
Credit: 305,637,084
RAC: 5,220
|
From what I see we did not find any Primes this year, but we will have to wait and see.
There are to many to review so I would not know myself.
This was fifty-fifty. The tested range was [4950146; 5372518]: the probability of 0 prime is 50.6% and of one prime or more 49.4%. Note that 4896418219 + 1 was found on 15 May. |
|
|
Michael GutierrezVolunteer moderator
Project administrator
Project scientist
Send message
Joined: 21 Mar 17
Posts: 341
ID: 764476
Credit: 46,281,510
RAC: 0
|
|
Cleanup Status:
May 30: 10797 tasks outstanding; 7924 affecting individual (271) scoring positions; 2638 affecting team (40) scoring positions.
May 31: 8973 tasks outstanding; 6000 affecting individual (262) scoring positions; 1937 affecting team (36) scoring positions.
Jun 1: 7079 tasks outstanding; 4400 affecting individual (248) scoring positions; 1436 affecting team (32) scoring positions.
Jun 2: 6030 tasks outstanding; 3168 affecting individual (238) scoring positions; 1185 affecting team (29) scoring positions.
Jun 3: 5323 tasks outstanding; 2697 affecting individual (225) scoring positions; 1042 affecting team (27) scoring positions.
Jun 4: 4601 tasks outstanding; 2281 affecting individual (217) scoring positions; 871 affecting team (24) scoring positions.
Jun 5: 3421 tasks outstanding; 1448 affecting individual (183) scoring positions; 234 affecting team (17) scoring positions.
Jun 6: 2708 tasks outstanding; 1022 affecting individual (160) scoring positions; 146 affecting team (12) scoring positions.
Jun 7: 2332 tasks outstanding; 803 affecting individual (149) scoring positions; 124 affecting team (11) scoring positions.
Jun 8: 1855 tasks outstanding; 608 affecting individual (130) scoring positions; 54 affecting team (9) scoring positions.
Jun 9: 1440 tasks outstanding; 438 affecting individual (104) scoring positions; 42 affecting team (8) scoring positions.
Jun 10: 1055 tasks outstanding; 247 affecting individual (77) scoring positions; 12 affecting team (7) scoring positions.
Jun 11: 866 tasks outstanding; 208 affecting individual (69) scoring positions; 5 affecting team (5) scoring positions.
Jun 12: 587 tasks outstanding; 137 affecting individual (52) scoring positions; 3 affecting team (3) scoring positions.
Jun 13: 481 tasks outstanding; 106 affecting individual (46) scoring positions; 3 affecting team (3) scoring positions.
Jun 14: 427 tasks outstanding; 89 affecting individual (41) scoring positions; 3 affecting team (3) scoring positions.
Jun 15: 359 tasks outstanding; 68 affecting individual (35) scoring positions; 0 affecting team (0) scoring positions.
Jun 16: 272 tasks outstanding; 41 affecting individual (23) scoring positions; 0 affecting team (0) scoring positions.
Jun 17: 231 tasks outstanding; 35 affecting individual (20) scoring positions; 0 affecting team (0) scoring positions.
Jun 18: 196 tasks outstanding; 33 affecting individual (19) scoring positions; 0 affecting team (0) scoring positions.
Jun 19: 168 tasks outstanding; 27 affecting individual (17) scoring positions; 0 affecting team (0) scoring positions.
Jun 20: 145 tasks outstanding; 24 affecting individual (16) scoring positions; 0 affecting team (0) scoring positions.
Jun 21: 107 tasks outstanding; 13 affecting individual (9) scoring positions; 0 affecting team (0) scoring positions.
Jun 22: 90 tasks outstanding; 11 affecting individual (7) scoring positions; 0 affecting team (0) scoring positions.
Jun 23: 72 tasks outstanding; 9 affecting individual (6) scoring positions; 0 affecting team (0) scoring positions.
Jun 24: 54 tasks outstanding; 5 affecting individual (5) scoring positions; 0 affecting team (0) scoring positions.
Jun 25: 44 tasks outstanding; 5 affecting individual (5) scoring positions; 0 affecting team (0) scoring positions.
Jun 26: 42 tasks outstanding; 5 affecting individual (5) scoring positions; 0 affecting team (0) scoring positions.
Jun 27: 33 tasks outstanding; 4 affecting individual (4) scoring positions; 0 affecting team (0) scoring positions.
Jun 28: 20 tasks outstanding; 2 affecting individual (2) scoring positions; 0 affecting team (0) scoring positions.
Jun 29: 20 tasks outstanding; 2 affecting individual (2) scoring positions; 0 affecting team (0) scoring positions.
____________
Time flies like an arrow. Fruit flies like a banana. |
|
|
Michael GutierrezVolunteer moderator
Project administrator
Project scientist
Send message
Joined: 21 Mar 17
Posts: 341
ID: 764476
Credit: 46,281,510
RAC: 0
|
|
Cleanup Status:
Jun 30: 17 tasks outstanding; 2 affecting individual (2) scoring positions; 0 affecting team (0) scoring positions.
Jul 1: 15 tasks outstanding; 2 affecting individual (2) scoring positions; 0 affecting team (0) scoring positions.
Jul 2: 13 tasks outstanding; 2 affecting individual (2) scoring positions; 0 affecting team (0) scoring positions.
Jul 3: 9 tasks outstanding; 1 affecting individual (1) scoring positions; 0 affecting team (0) scoring positions.
____________
Time flies like an arrow. Fruit flies like a banana. |
|
|
Michael GutierrezVolunteer moderator
Project administrator
Project scientist
Send message
Joined: 21 Mar 17
Posts: 341
ID: 764476
Credit: 46,281,510
RAC: 0
|
|
The results are final!
During the 5 day challenge, we completed 117,428 tasks. Unfortunately no new GFN-19 primes were found during the challenge, although the inimitable tng did find one days before, presumably while 'warming up', so I'm counting that as a W! The challenge boosted the project's average output to over 50 times typical at its peak!
95 teams and 444 individuals participated in the Challenge.
Top Three Individuals:
1. [TA]Skillz
2. vaughan
3. tng
Top Three Teams:
1. TeAm AnandTech
2. Antarctic Crunchers
3. SETI.Germany
I was very entertained to notice that all 6 of the Top Ranks are identical to those from last year's Once In a Blue Moon Challenge (save for skillz overtaking vaughan!). Made it a lot easier to copy/paste this announcement as well lol ;-)
Up next is the Pi Approximation Day challenge, starting in 2 weeks! Hasta la justa!
____________
Time flies like an arrow. Fruit flies like a banana. |
|
|
