## Other

drummers-lowrise

Message boards : Number crunching : Leonids Shower Challenge

 Subscribe SortOldest firstNewest firstHighest rated posts first
Author Message
Roger
Volunteer developer
Volunteer tester

Joined: 27 Nov 11
Posts: 1138
ID: 120786
Credit: 268,621,444
RAC: 0

Message 122203 - Posted: 7 Nov 2018 | 12:35:16 UTC

Welcome to the Leonids Shower Challenge

The Leonids is an average shower, producing up to 15 meteors per hour at its peak. This shower is unique in that it has a cyclonic peak about every 33 years where hundreds of meteors per hour can be seen. That last of these occurred in 2001. The Leonids is produced by dust grains left behind by comet Tempel-Tuttle, which was discovered in 1865. The shower runs annually from November 6-30. It peaks this year on the night of the 17th and morning of the 18th. The waxing gibbous moon will set shortly after midnight leaving fairly dark skies for what could be a good early morning show. Best viewing will be from a dark location after midnight. Meteors will radiate from the constellation Leo, but can appear anywhere in the sky.

Arithmetic progressions
In mathematics, an arithmetic progression (AP) or arithmetic sequence is a sequence of numbers such that the difference between the consecutive terms is constant. For instance, the sequence 5, 7, 9, 11, 13, 15 … is an arithmetic progression with common difference of 2.

An arithmetic progression of primes is a sequence of primes with a common difference between any two successive numbers in the sequence. For example 3, 7, 11 is an arithmetic progression of 3 primes with a common difference of 4.

For an arithmetic progression (AP) of primes, AP-k is k primes of the form p + d*n for some d (the common difference between the primes) and k consecutive values of n. The above AP-3 is 3 + 4*n for n=0,1,2.

How to Participate?
To participate in the Challenge, please select only the AP 27 (AP27) project in your PrimeGrid preferences section. The challenge will begin 17th November 2018 12:00 UTC and end 20th November 2018 12:00 UTC. Application builds are available for Linux , Windows and MacIntel CPUs and GPUs. CPU apps are only available for 64 bit CPUs. High end Nvidia GPUs will have a very large advantage.

Tasks will take ~29 hours on average for CPUs and ~75 minutes on average for GPUs. If your computer is highly overclocked, please consider "stress testing" it. If you haven't run the AP app before, we strongly suggest running it before the challenge while you are monitoring the temperatures. You don't want to turn your machine into a meteor!

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 tasks issued AFTER 17 November 2018 12:00 UTC and received BEFORE 20 November 2018 12:00 UTC will be considered for credit. We will be using the same scoring method as we currently use for BOINC credits.
A quorum of 2 is NOT needed to award Challenge score - i.e. no double checker. Therefore, each returned result will earn a Challenge score. Please note that if the result is eventually declared invalid, the score will be removed.

At the Conclusion of the Challenge

ABORTING tasks allows them to be recycled immediately; thus a much faster "clean up" to the end of a Challenge. DETACHING, RESETTING, and PAUSING tasks causes them to remain in limbo until they EXPIRE. Therefore, we must wait until tasks expire to send them out to be completed.

Please consider either completing what's in the queue or ABORTING them. Thank you. :)

- AP 27 Search
- AP 26 Search
- http://en.wikipedia.org/wiki/Arithmetic_progression
- Jens Kruse Andersen's Primes in Arithmetic Progression Records
____________

FishFry

Joined: 30 Nov 16
Posts: 27
ID: 472934
Credit: 515,213,055
RAC: 0

Message 122481 - Posted: 14 Nov 2018 | 20:41:56 UTC - in response to Message 122203.

Tasks will take ~29 hours on average for CPUs and ~75 minutes on average for GPUs.

I never ran AP27 on CPUs. Anyone know if this application works well with multi-threading like LLR?

Good luck everyone!

Rick Reynolds

Joined: 13 Aug 10
Posts: 8159
ID: 65483
Credit: 6,713,024,615
RAC: 4,705

Message 122483 - Posted: 14 Nov 2018 | 20:43:21 UTC - in response to Message 122481.

Tasks will take ~29 hours on average for CPUs and ~75 minutes on average for GPUs.

I never ran AP27 on CPUs. Anyone know if this application works well with multi-threading like LLR?

Good luck everyone!

And if you are interested here's a thread on cpu times posted by a few folks who tested.

FishFry

Joined: 30 Nov 16
Posts: 27
ID: 472934
Credit: 515,213,055
RAC: 0

Message 122484 - Posted: 14 Nov 2018 | 20:49:34 UTC - in response to Message 122483.

Thanks Rick!

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 122485 - Posted: 14 Nov 2018 | 20:51:24 UTC - in response to Message 122483.

Tasks will take ~29 hours on average for CPUs and ~75 minutes on average for GPUs.

I never ran AP27 on CPUs. Anyone know if this application works well with multi-threading like LLR?

Good luck everyone!

And if you are interested here's a thread on cpu times posted by a few folks who tested.

The AP27 CPU app is NOT multithreaded.
____________
My lucky number is 75898524288+1

k4m1k4z3

Joined: 26 Apr 10
Posts: 26
ID: 59440
Credit: 1,127,811,157
RAC: 162,646

Message 122519 - Posted: 15 Nov 2018 | 17:59:42 UTC

Are the AP27 GPU tasks fairly CPU dependent? I am trying to understand why my GTX1080 paired with i7 6700k is much faster than GTX1080 paired with Ryzen 1700 (1200 seconds vs 2000 seconds) perhaps there are other factors slowing me down...

Eudy Silva

Joined: 26 Aug 17
Posts: 1607
ID: 918937
Credit: 354,081,280
RAC: 228,757

Message 122520 - Posted: 15 Nov 2018 | 18:10:47 UTC - in response to Message 122519.

You're using driver version 41634 on the machine with the GTX 1080 / AMD Ryzen.
The 4xxx NVidia driver series made my GTX 1060 3G about 30% slower when crunching AP27.
Try driver 39924 on your machine to see if solves the slowness.

Edit:
See these
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment.

k4m1k4z3

Joined: 26 Apr 10
Posts: 26
ID: 59440
Credit: 1,127,811,157
RAC: 162,646

Message 122523 - Posted: 15 Nov 2018 | 18:55:53 UTC

399 drivers helped a lot! Down to 1100 seconds (this card is clocked higher than the one in the i7 machine)

T-Armstrong

Joined: 3 Aug 10
Posts: 275
ID: 64825
Credit: 1,273,774,823
RAC: 851,526

Message 122535 - Posted: 16 Nov 2018 | 0:48:23 UTC

Where is my Post ?

I:m Suspend ?

is not Fair Play

____________

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 122536 - Posted: 16 Nov 2018 | 0:51:14 UTC - in response to Message 122535.

Where is my Post ?

I:m Suspend ?

No.
____________
My lucky number is 75898524288+1

T-Armstrong

Joined: 3 Aug 10
Posts: 275
ID: 64825
Credit: 1,273,774,823
RAC: 851,526

Message 122537 - Posted: 16 Nov 2018 | 1:01:05 UTC - in response to Message 122536.

ok, thanks

understand

____________

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 122569 - Posted: 17 Nov 2018 | 11:46:42 UTC

What an auspicious start. It's a little early for the challenge, but we just found an AP26!

More details shortly. When I'm fully awake. After I hit the GO button on my crunchers when the challenge starts!

____________
My lucky number is 75898524288+1

gemini8

Joined: 2 Jan 16
Posts: 101
ID: 434794
Credit: 213,827,289
RAC: 152,240

Message 122575 - Posted: 17 Nov 2018 | 12:29:10 UTC

Off we go. :-)
First results were in after less than a quarter of an hour.
____________
Greetings, Jens

1461*2^3373383+1

SEARCHER

Joined: 15 Jul 08
Posts: 101
ID: 25629
Credit: 1,702,368,904
RAC: 6,307,927

Message 122587 - Posted: 17 Nov 2018 | 13:26:25 UTC

+ 1 Cruncher more
____________
Member of Charity Team

Overtonesinger

Joined: 8 Apr 12
Posts: 3
ID: 137853
Credit: 71,637,047
RAC: 81,302

Message 122618 - Posted: 18 Nov 2018 | 8:59:23 UTC - in response to Message 122520.

I have a Ryzen (1800X) too ... + Asus(R.O.G.) GTX 1070 ......
- and was also wondering the same thing: WHY so slow?

Rolling back to 399.24 driver immediatelly!

Thanx a lot! :-)
____________

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 122624 - Posted: 18 Nov 2018 | 12:54:17 UTC

After the first day:

Challenge: Leonids Shower
App: 11 (AP27)
(As of 2018-11-18 12:53:04 UTC)

72175 tasks have been sent out. [CPU/GPU/anonymous_platform: 8916 (12%) / 63259 (88%) / 0 (0%)]

Of those tasks that have been sent out:

4755 (7%) came back with some kind of an error. [337 (0%) / 4418 (6%) / 0 (0%)]
40281 (56%) have returned a successful result. [1584 (2%) / 38697 (54%) / 0 (0%)]
27142 (38%) are still in progress. [6995 (10%) / 20147 (28%) / 0 (0%)]

Of the tasks that have been returned successfully:

16088 (40%) are pending validation. [660 (2%) / 15428 (38%) / 0 (0%)]
24189 (60%) have been successfully validated. [924 (2%) / 23265 (58%) / 0 (0%)]
2 (0%) were invalid. [0 (0%) / 2 (0%) / 0 (0%)]
2 (0%) are inconclusive. [0 (0%) / 2 (0%) / 0 (0%)]
____________
My lucky number is 75898524288+1

Grebuloner
Volunteer tester

Joined: 2 Nov 09
Posts: 402
ID: 49572
Credit: 2,195,704,211
RAC: 915,929

Message 122636 - Posted: 18 Nov 2018 | 17:58:42 UTC

Well, it appears that a good percentage of those errors were mine. Had a generally unmonitored host running PPS-Sieve for a long time so didn't notice that OpenCL got borked in a windows 10 update. Took a few driver reinstalls and reboots and it's back on track. Just wish I had checked it out sooner so as to not miss so much credit :(.
____________
Eating more cheese on Thursdays.

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 122639 - Posted: 18 Nov 2018 | 18:26:31 UTC

APs found so far during the challenge. To be in this list, the AP's finder had to return the task during the challenge, and it has to have also been validated by the double checker. The double check could happen after the challenge; what counts here is when the finder returned the task.

+-----+------+ | len | cnt | +-----+------+ | 20 | 1977 | | 21 | 488 | | 22 | 117 | | 23 | 30 | | 24 | 5 | +-----+------+

I don't have information on APs of length 24 or below that have been returned but are awaiting wingmen for validation, but there's undoubtedly a bunch of them. There is one pending AP25 waiting for a wingman.
____________
My lucky number is 75898524288+1

Grebuloner
Volunteer tester

Joined: 2 Nov 09
Posts: 402
ID: 49572
Credit: 2,195,704,211
RAC: 915,929

Message 122642 - Posted: 18 Nov 2018 | 19:42:16 UTC

Between my own results and seeing it in the results you've posted, I've always been fascinated about the ~4:1 ratio of AP(n):AP(n+1) finds. Is there a known mathematical reason for this? I know there are an infinite number of APs, but I haven't come across any quantitative information in my searches.
____________
Eating more cheese on Thursdays.

JeppeSN

Joined: 5 Apr 14
Posts: 1563
ID: 306875
Credit: 36,252,500
RAC: 8,726

Message 122644 - Posted: 18 Nov 2018 | 20:06:30 UTC - in response to Message 122642.

One could try a semi-log plot and see if the points are on a line, and if the slope is consistent with that 1/4.0 hypothesis. /JeppeSN

Dave

Joined: 13 Feb 12
Posts: 2950
ID: 130544
Credit: 1,750,422,384
RAC: 3,203,715

Message 122646 - Posted: 18 Nov 2018 | 20:25:11 UTC

Len # Ratio: 20 2702 21 705 3.832624113 22 156 4.519230769 23 31 5.032258065 24 8 3.875 25 2 4

robish
Volunteer moderator
Volunteer tester

Joined: 7 Jan 12
Posts: 2029
ID: 126266
Credit: 6,085,663,747
RAC: 3,362,305

Message 122647 - Posted: 18 Nov 2018 | 20:25:30 UTC - in response to Message 122644.

One could try a semi-log plot and see if the points are on a line, and if the slope is consistent with that 1/4.0 hypothesis. /JeppeSN

Interesting. Worth spending some time on I'd say.
____________
My lucky numbers 10590941048576+1 and 224584605939537911+81292139*23#*n for n=0..26

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 122667 - Posted: 19 Nov 2018 | 13:01:40 UTC

Day 2L

Challenge: Leonids Shower
App: 11 (AP27)
(As of 2018-11-19 12:58:31 UTC)

126686 tasks have been sent out. [CPU/GPU/anonymous_platform: 14390 (11%) / 112296 (89%) / 0 (0%)]

Of those tasks that have been sent out:

8773 (7%) came back with some kind of an error. [766 (1%) / 8007 (6%) / 0 (0%)]
86966 (69%) have returned a successful result. [5480 (4%) / 81487 (64%) / 0 (0%)]
30948 (24%) are still in progress. [8144 (6%) / 22803 (18%) / 0 (0%)]

Of the tasks that have been returned successfully:

21984 (25%) are pending validation. [1388 (2%) / 20597 (24%) / 0 (0%)]
64977 (75%) have been successfully validated. [4092 (5%) / 60885 (70%) / 0 (0%)]
5 (0%) were invalid. [0 (0%) / 5 (0%) / 0 (0%)]
0 (0%) are inconclusive. [0 (0%) / 0 (0%) / 0 (0%)]
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 122714 - Posted: 20 Nov 2018 | 3:12:15 UTC

With less than a day remaining in the challenge, it's time to remind everyone...

At the Conclusion of the Challenge

We would prefer users "moving on" to finish those tasks they have downloaded, if not then please ABORT the WU's (and then UPDATE the PrimeGrid project) instead of DETACHING, RESETTING, or PAUSING.

ABORTING WU's allows them to be recycled immediately; thus a much faster "clean up" to the end of 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. Thank you!
____________
My lucky number is 75898524288+1

River~~

Joined: 17 Mar 07
Posts: 342
ID: 6533
Credit: 15,792,075
RAC: 0

Message 122727 - Posted: 20 Nov 2018 | 11:38:52 UTC

Hi Leonidists!

With significantly less than an hour to go for your work to qualify, are you biting your nails with a job that is predicted to end very close to the deadline?

Won't you be annoyed if the work complets at 1155 UT and doesn't upload till 1257? !!????!!!!!

Normal BOINC client behaviour is to wait for an hour before uploading, so that if other work finishes in that hour they can all be uploaded together. This saves net traffic and load on the servers. But very annoying when time is critical.

There is a fix!

Well actually there are four fixes -- one is the manual one of being at your computer at 1157 to update PG, and the second is to write a BASH or BAT script to do the same. But those are workarounds rather than fixes, and there are more elegant fixes built into the client itself.

You can reduce this 1 hour wait to just 1 minute either by using a startup switch or a tweak in the config file.

I recommend using the file, but please do remember to set the option to zero afterwards, to save that server load, please.

See this BOINC wiki page for details of both methods
____________
My computers found:

9831*21441403+1 is a quadhectokilo prime prime, ie >400,000 digits ;)

252031090528237591 + 65521*149*23*19*17*13*11*7*5*3*2*n is prime for every n in { 0..20 } (an arithemtic progression of 21 primes)

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 122728 - Posted: 20 Nov 2018 | 12:22:28 UTC - in response to Message 122727.

Hi Leonidists!

With significantly less than an hour to go for your work to qualify, are you biting your nails with a job that is predicted to end very close to the deadline?

I had one CPU task that I knew was going to be really close.

It ended up finishing at 11:59:59 and being reported at 12:00:03. Oh well.
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 122729 - Posted: 20 Nov 2018 | 12:27:20 UTC

Final totals:

Challenge: Leonids Shower
App: 11 (AP27)
(As of 2018-11-20 12:19:40 UTC)

172928 tasks have been sent out. [CPU/GPU/anonymous_platform: 17659 (10%) / 155269 (90%) / 0 (0%)]

Of those tasks that have been sent out:

12719 (7%) came back with some kind of an error. [1657 (1%) / 11062 (6%) / 0 (0%)]
133916 (77%) have returned a successful result. [10047 (6%) / 123869 (72%) / 0 (0%)]
25544 (15%) are still in progress. [5800 (3%) / 19744 (11%) / 0 (0%)]

Of the tasks that have been returned successfully:

20513 (15%) are pending validation. [1255 (1%) / 19258 (14%) / 0 (0%)]
113396 (85%) have been successfully validated. [8792 (7%) / 104604 (78%) / 0 (0%)]
7 (0%) were invalid. [0 (0%) / 7 (0%) / 0 (0%)]
0 (0%) are inconclusive. [0 (0%) / 0 (0%) / 0 (0%)]

Some totals on the APs found:

+-----+------+ | len | cnt | +-----+------+ | 20 | 7068 | | 21 | 1710 | | 22 | 378 | | 23 | 107 | | 24 | 19 | | 25 | 4 | +-----+------+

As before, those were discovered during the challenge and have been validated, rather than having been part of a challenge task. There undoubtedly more unvalidated small (length 24 or shorter) APs waiting for wingmen.
____________
My lucky number is 75898524288+1

robish
Volunteer moderator
Volunteer tester

Joined: 7 Jan 12
Posts: 2029
ID: 126266
Credit: 6,085,663,747
RAC: 3,362,305

Message 122730 - Posted: 20 Nov 2018 | 12:30:38 UTC

Well done dh1saj, you passed me out in the last few minutes :)

Good chase :)
____________
My lucky numbers 10590941048576+1 and 224584605939537911+81292139*23#*n for n=0..26

Roger
Volunteer developer
Volunteer tester

Joined: 27 Nov 11
Posts: 1138
ID: 120786
Credit: 268,621,444
RAC: 0

Message 122733 - Posted: 20 Nov 2018 | 12:37:25 UTC

[H]atp1916 from TeAm AnandTech smashed it with 26,356,317 provisional points! More than total current credit showing on your AP badge, and AP24 discovered.
Street Creed to TeAm AnandTech with 86,625,318. Truly Beastful.

Better bring it to the next Challenge Brah

Remember to vote on next year's challenges: 2019 Challenge Voting

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 122738 - Posted: 20 Nov 2018 | 13:16:18 UTC

The cleanup begins! I expect the cleanup to take 3 to 6 weeks.

Leonids Cleanup:
Nov 20: Leonids Shower: 19854 tasks outstanding; 18650 affecting individual (296) scoring positions; 9307 affecting team (75) scoring positions.
Nov 21: Leonids Shower: 12899 tasks outstanding; 10631 affecting individual (287) scoring positions; 5614 affecting team (61) scoring positions.
Nov 22: Leonids Shower: 11024 tasks outstanding; 8846 affecting individual (280) scoring positions; 4817 affecting team (60) scoring positions.
Nov 23: Leonids Shower: 10593 tasks outstanding; 8510 affecting individual (278) scoring positions; 4638 affecting team (60) scoring positions.
Nov 24: Leonids Shower: 10265 tasks outstanding; 8141 affecting individual (275) scoring positions; 4502 affecting team (60) scoring positions.
Nov 25: Leonids Shower: 10015 tasks outstanding; 7893 affecting individual (272) scoring positions; 4402 affecting team (59) scoring positions.
Nov 26: Leonids Shower: 7133 tasks outstanding; 5041 affecting individual (247) scoring positions; 2191 affecting team (49) scoring positions.
Nov 27: Leonids Shower: 3549 tasks outstanding; 1843 affecting individual (181) scoring positions; 606 affecting team (33) scoring positions.
Nov 28: Leonids Shower: 1423 tasks outstanding; 500 affecting individual (116) scoring positions; 78 affecting team (15) scoring positions.
Nov 29: Leonids Shower: 699 tasks outstanding; 189 affecting individual (69) scoring positions; 8 affecting team (6) scoring positions.
Nov 30: Leonids Shower: 421 tasks outstanding; 100 affecting individual (47) scoring positions; 4 affecting team (3) scoring positions.
Dec 1: Leonids Shower: 261 tasks outstanding; 61 affecting individual (34) scoring positions; 3 affecting team (2) scoring positions.
Dec 2: Leonids Shower: 183 tasks outstanding; 50 affecting individual (30) scoring positions; 3 affecting team (2) scoring positions.
Dec 3: Leonids Shower: 168 tasks outstanding; 47 affecting individual (30) scoring positions; 3 affecting team (2) scoring positions.
Dec 4: Leonids Shower: 124 tasks outstanding; 37 affecting individual (22) scoring positions; 3 affecting team (2) scoring positions.
Dec 5: Leonids Shower: 73 tasks outstanding; 17 affecting individual (10) scoring positions; 0 affecting team (0) scoring positions.
Dec 6: Leonids Shower: 30 tasks outstanding; 3 affecting individual (2) scoring positions; 0 affecting team (0) scoring positions.
____________
My lucky number is 75898524288+1

Roger
Volunteer developer
Volunteer tester

Joined: 27 Nov 11
Posts: 1138
ID: 120786
Credit: 268,621,444
RAC: 0

Message 123196 - Posted: 6 Dec 2018 | 21:45:46 UTC

The results are final!

Top 3 individuals:

1: [H]atp1916
2: vaughan
3: tng*

Top 3 teams:

1: TeAm AnandTech
3: Czech National Team

Congratulations to the winners, and well done to everyone who participated.
See you at the Winter Solstice Challenge!
____________

Eudy Silva

Joined: 26 Aug 17
Posts: 1607
ID: 918937
Credit: 354,081,280
RAC: 228,757

Message 124846 - Posted: 15 Jan 2019 | 5:35:42 UTC

I'm reviewing my software tool that generates the overall stats for the challenge series in each year.
I'm getting some errors.
Among them, one happens when redoing the overall 2018 stats.
Challenge Leonids Shower results seems to have an inconsistence in the individual vs team results.

The individual results for the Leonids Shower challenge list user Tejas (rank 280) of team Hagerhill KY (teamid=4548) with a score of 92,989 points.
But in the team results of the same challenge, team Hagerhill KY is not listed, although Tejas scored 92,989 for it and this team would be ranked 64th.
Why isn't it there ?
Am I missing something ? (it happens pretty often :) )
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment.

JeppeSN

Joined: 5 Apr 14
Posts: 1563
ID: 306875
Credit: 36,252,500
RAC: 8,726

Message 124850 - Posted: 15 Jan 2019 | 6:17:57 UTC - in response to Message 124846.

The individual results for the Leonids Shower challenge list user Tejas (rank 280) of team Hagerhill KY (teamid=4548) with a score of 92,989 points.
But in the team results of the same challenge, team Hagerhill KY is not listed, although Tejas scored 92,989 for it and this team would be ranked 64th.
Why isn't it there ?

That user may have changed teams many times. Today he seems to be in "The Scottish Boinc Team".

What happens if a user gets a task while in one team (or when outside a team), but is in a different team (or outside teams) when he returns the result? When the challenge ends, he might be in a third team (and it could be that he returns no results (in due time) for his last team). /JeppeSN

Roger
Volunteer developer
Volunteer tester

Joined: 27 Nov 11
Posts: 1138
ID: 120786
Credit: 268,621,444
RAC: 0

Message 124851 - Posted: 15 Jan 2019 | 6:24:05 UTC

I note that Tejas is no longer a member of Hagerhill KY. He joined The Scottish Boinc Team. Maybe the User stats don't get changed when you change teams after the fact, but Team stats do?
Users can also change teams during a Challenge, which creates all kinds of headaches for the scoreboards. As long as the treatment is consistent I don't think it matters a great deal.

Eudy Silva

Joined: 26 Aug 17
Posts: 1607
ID: 918937
Credit: 354,081,280
RAC: 228,757

Message 124867 - Posted: 15 Jan 2019 | 14:00:56 UTC - in response to Message 124851.

I'm aware of the user changing teams during the challenge.
(A user might even change his own name during the challenge and that's why I work with the user id, instead of relying on his/her name.)
Regarding a user changing teams during a challenge, I just hadn't realized the impact it could have when calculating the user rank within his team for a challenge, in some specific cases.

The problem I'm having is to calculate Tejas' rank within his team in the 8th challenge of 2018 (Leonids Shower), since his team is not listed in the challenge's teams results.
One explanation I can think of is that Tejas crunched for some other team during the challenge (say, AnandTech) and after crunching the last task, he switched to team Hagerhill KY.
As nobody else in team Hagerhill KY crunched in the challenge, Hagerhill KY didn't score in that challenge.
The individual stats for the challenge lists Tejas with his points in the challenge and the current team he/she was in at the time the stats where generated.
That would explain Hagerhill KY not being listed as a team with score.

Guess that in cases like this, I'll just consider it not possible to calculate the user rank within his team in the challenge and report this as "rank in team = (--)".

I'm not sure why I didn't stumble in this before (I mean, getting this as an error in my tool).

So, the overall individual stats for 2018 is not 100% correct, since it shows Tejas ranked 1st in his team in the Leonids Shower challenge.
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment.

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13679
ID: 53948
Credit: 304,001,596
RAC: 232,729

Message 124868 - Posted: 15 Jan 2019 | 14:16:19 UTC

Team scores are tabulated on a task by task basis. Each individual tasks contains a teamid as well as a userid. I'm fairly sure the teamid is set when the task is sent to a computer. If a user switches teams during the challenge, some of his tasks will be for different teams.

Worse yet, there's been cases where a team was deleted during a challenge. That actually broke the official challenge results once.

So, yeah, there's some strange stuff that can happen with the statistics.
____________
My lucky number is 75898524288+1

Eudy Silva

Joined: 26 Aug 17
Posts: 1607
ID: 918937
Credit: 354,081,280
RAC: 228,757

Message 124870 - Posted: 15 Jan 2019 | 14:22:24 UTC

Thank you for the input on this issue, guys !
____________
"Accidit in puncto, quod non contingit in anno."
Something that does not occur in a year may, perchance, happen in a moment.

Message boards : Number crunching : Leonids Shower Challenge