## Other

drummers-lowrise

Message boards : Number crunching : Wallis is Born Challenge

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

Joined: 27 Nov 11
Posts: 1137
ID: 120786
Credit: 266,956,799
RAC: 7,867

Message 100765 - Posted: 10 Nov 2016 | 13:08:57 UTC

Welcome to the Wallis is Born Challenge

John Wallis (23 November 1616 – 28 October 1703) was an English mathematician who is given partial credit for the development of infinitesimal calculus. Between 1643 and 1689 he served as chief cryptographer for Parliament and, later, the royal court. He is credited with introducing the symbol ∞ for infinity. He similarly used 1/∞ for an infinitesimal. Asteroid 31982 Johnwallis was named after him.

Wallis made significant contributions to trigonometry, calculus, geometry, and the analysis of infinite series. In his Opera Mathematica I (1695) he introduced the term "continued fraction".

Wallis rejected as absurd the now usual idea of a negative number as being less than nothing, but accepted the view that it is something greater than infinity. (The argument that negative numbers are greater than infinity involves the quotient 1/x and considering what happens as x approaches and then crosses the point x = 0 from the positive side.) Despite this he is generally credited as the originator of the idea of the number line, in which numbers are represented geometrically in a line with the negative numbers represented by lengths opposite in direction to lengths of positive numbers.

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 18th November 2016 18:00 UTC and end 23rd November 2016 18:00 UTC. Application builds are available for Linux , Windows and MacIntel CPU and GPUs. High end Nvidia GPUs will have a very large advantage.

Tasks will take ~23 hours on average for CPUs and ~100 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.

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 18 November 2016 18:00 UTC and received BEFORE 23 November 2016 18:00 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

We kindly ask users "moving on" to ABORT their tasks instead of DETACHING, RESETTING, or PAUSING.

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

____________

[B@P] Daniel

Joined: 18 Dec 11
Posts: 70
ID: 123596
Credit: 638,082,401
RAC: 0

Message 100771 - Posted: 10 Nov 2016 | 17:02:36 UTC - in response to Message 100765.

What about HyperThreading? Disable it or leave it enabled?
____________

Anthony Ayiomamitis

Joined: 19 Mar 15
Posts: 2707
ID: 386066
Credit: 745,841,621
RAC: 306,121

Message 100773 - Posted: 10 Nov 2016 | 17:25:37 UTC - in response to Message 100771.
Last modified: 10 Nov 2016 | 17:31:22 UTC

What about HyperThreading? Disable it or leave it enabled?

You are fine with HT on.

Some CPU- and GPU-based results:

i7 6700k - 23.46 hrs per unit (8/8 virtual cores)
i3 6300 - 24.50 hrs per unit (4/4 virtual cores)
i3 6100 - 24.57 hrs per unit (4/4 virtual cores)

i7 3630QM - 31.94 hrs per unit (6/8 virtual cores)

GTX 660M - 7.12 hrs per unit

Crun-chi
Volunteer tester

Joined: 25 Nov 09
Posts: 2976
ID: 50683
Credit: 57,167,716
RAC: 56,558

Message 100778 - Posted: 10 Nov 2016 | 18:52:53 UTC - in response to Message 100773.
Last modified: 10 Nov 2016 | 18:53:18 UTC

GTX 750 Ti - 7100 sec per WU
____________
314187728^131072+1 GENERALIZED FERMAT :)
93*10^1029523-1 REPDIGIT PRIME
31*332^367560+1 CRUS PRIME
Proud member of team Aggie The Pew. Go Aggie!

[B@P] Daniel

Joined: 18 Dec 11
Posts: 70
ID: 123596
Credit: 638,082,401
RAC: 0

Message 100779 - Posted: 10 Nov 2016 | 19:22:10 UTC - in response to Message 100773.

What about HyperThreading? Disable it or leave it enabled?

You are fine with HT on.

Thanks.

GTX 1070 - 1542 secs per WU (about 26 minutes)
____________

[B@P] Daniel

Joined: 18 Dec 11
Posts: 70
ID: 123596
Credit: 638,082,401
RAC: 0

Message 100782 - Posted: 10 Nov 2016 | 20:35:43 UTC

Durations for new WUs:
GTX 1070 - ~1700 secs (~28 mins)
GTX 970 - ~2800 secs (~47 mins)
____________

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 100788 - Posted: 10 Nov 2016 | 23:42:23 UTC

GTX 580: 4630 seconds
____________
My lucky number is 75898524288+1

JeppeSN

Joined: 5 Apr 14
Posts: 1266
ID: 306875
Credit: 17,694,119
RAC: 35,588

Message 100790 - Posted: 10 Nov 2016 | 23:48:18 UTC

I note the following:

Wallis's birthday can be given as either 23 November 1616 or 3 December 1616.

In that period, England was not yet using the modern ("new style") calendar, the Gregorianske calendar, that is universal today. So at the day of his birth, it was 23 Nov in England and many other parts of the world, but 3 Dec according to other areas where they had switched to the new calendar.

For this reason, the 400th birthday of Wallis could be celebrated on either 23 Now or 3 Dec.

/JeppeSN

Brook Harste

Joined: 30 Nov 08
Posts: 1345
ID: 32401
Credit: 805,483,198
RAC: 375

Message 100793 - Posted: 11 Nov 2016 | 5:18:09 UTC

GTX 980Ti 1,487 seconds
GTX 770 5,760 seconds
____________

Dave

Joined: 13 Feb 12
Posts: 2779
ID: 130544
Credit: 931,529,915
RAC: 340,866

Message 100795 - Posted: 11 Nov 2016 | 6:31:28 UTC - in response to Message 100790.

I note the following:

Wallis's birthday can be given as either 23 November 1616 or 3 December 1616.

In that period, England was not yet using the modern ("new style") calendar, the Gregorianske calendar, that is universal today. So at the day of his birth, it was 23 Nov in England and many other parts of the world, but 3 Dec according to other areas where they had switched to the new calendar.

For this reason, the 400th birthday of Wallis could be celebrated on either 23 Now or 3 Dec.

/JeppeSN

Good point!

Although the calendar changes were in 1582 weren't they?

My GPU times are all over the place - will post times when I can benchmark properly.

JeppeSN

Joined: 5 Apr 14
Posts: 1266
ID: 306875
Credit: 17,694,119
RAC: 35,588

Message 100798 - Posted: 11 Nov 2016 | 8:59:05 UTC - in response to Message 100795.

I note the following:

Wallis's birthday can be given as either 23 November 1616 or 3 December 1616.

In that period, England was not yet using the modern ("new style") calendar, the Gregorianske calendar, that is universal today. So at the day of his birth, it was 23 Nov in England and many other parts of the world, but 3 Dec according to other areas where they had switched to the new calendar.

For this reason, the 400th birthday of Wallis could be celebrated on either 23 Now or 3 Dec.

/JeppeSN

Good point!

Although the calendar changes were in 1582 weren't they?

My GPU times are all over the place - will post times when I can benchmark properly.

Yes, the changes came in 1582, but only in Southern Europe (and its colonies). Many areas adopted the new system much later. So during the life of Wallis, two competing calendar systems existed. Which is why I mentioned it.

England, where Wallis was born and lived, did not follow the Pope as closely as some of the southern countries of Europe, and they stuck to the old calendar. Not before 1752, long after Wallis's death, would England (then part of the United Kingdom) make the switch.

/JeppeSN

River~~

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

Message 100799 - Posted: 11 Nov 2016 | 10:10:20 UTC - in response to Message 100790.

...
the 400th birthday of Wallis could be celebrated on either 23 Now or 3 Dec.
...
/JeppeSN

hi,

It would be twice as appropriate to run the challenge from 23 Nov to 3 Dec.

And twice the length of the challenge, too :)

River~~

River~~

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

Message 100821 - Posted: 11 Nov 2016 | 15:42:47 UTC

Been testing the AP27 app on some of my computers.

Am I right that it is not designed to run on 32bit? The scheduler refuses to give my 32 bit boxes any AP work, but doesn't say why.

I have also noticed that suspend-in-memory has been implemented differently to how it was done in LLR -- when an AP task is suspended the Linux command "ps" lists its state as "S" (interruptible sleep), whereas LLR shows up as "T" (stopped by job control). This difference makes no practical difference to me as a user, it was just unexpected.

Also ps lists the AP apps as being threaded, whereas LLR are not. Run times suggest that the threads do not run concurrently, or at least not to a measurable degree.

None of my boxes seem to complain about getting too hot, and all the run times are under five days so hopefully all my 64bit machines will contribute at least one WU per core/HT thread.

R~~

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 100826 - Posted: 11 Nov 2016 | 16:16:14 UTC - in response to Message 100821.

Been testing the AP27 app on some of my computers.

Am I right that it is not designed to run on 32bit? The scheduler refuses to give my 32 bit boxes any AP work, but doesn't say why.

I have also noticed that suspend-in-memory has been implemented differently to how it was done in LLR -- when an AP task is suspended the Linux command "ps" lists its state as "S" (interruptible sleep), whereas LLR shows up as "T" (stopped by job control). This difference makes no practical difference to me as a user, it was just unexpected.

Also ps lists the AP apps as being threaded, whereas LLR are not. Run times suggest that the threads do not run concurrently, or at least not to a measurable degree.

None of my boxes seem to complain about getting too hot, and all the run times are under five days so hopefully all my 64bit machines will contribute at least one WU per core/HT thread.

R~~

You are correct; there is no 32 bit app. If you look at the description of the project (where you select the project) you'll see that there's no 32 bit apps available.

As for the task suspension, that's done by BOINC itself, not the app. BOINC doesn't tell the app "go to sleep". BOINC tells the operating system "Put that process to sleep". If you're seeing a difference in behavior, it's probably because LLR is actually two processes, a wrapper and the LLR program, whereas AP27 doesn't use a wrapper so it's just one process.

If the app is listed as being "threaded", it's probably because it's linked to the multithreaded versions of the libraries. That's not needed (it's not actually a multithreaded app), but it won't hurt anything. It might slow down some system calls by an insignificant amount, but won't affect the calculation speed.

____________
My lucky number is 75898524288+1

Skivelitis2

Joined: 28 Mar 15
Posts: 42
ID: 388656
Credit: 283,325,222
RAC: 57,699

Message 100833 - Posted: 11 Nov 2016 | 23:12:05 UTC
Last modified: 11 Nov 2016 | 23:16:30 UTC

Probably not the place to ask but...Will GPU apps work on a WinXP 32-bit host?

Edit: Found the thread on supported systems.

stream
Volunteer moderator
Volunteer developer
Volunteer tester

Joined: 1 Mar 14
Posts: 790
ID: 301928
Credit: 483,980,905
RAC: 147,457

Message 100835 - Posted: 11 Nov 2016 | 23:19:20 UTC - in response to Message 100821.

Am I right that it is not designed to run on 32bit? The scheduler refuses to give my 32 bit boxes any AP work, but doesn't say why.

As far as I remember 32-bit CPU app was significantly slower then 64-bit so it wasn't enabled.

As for GPU app, it was Michael's decision to not send it to 32-bit systems although GPU app itself is 32-bit, motivating that number of such systems (32-bit OS and capable GPU with 2+ GB of video RAM) was very low.

I have also noticed that suspend-in-memory has been implemented differently to how it was done in LLR -- when an AP task is suspended the Linux command "ps" lists its state as "S" (interruptible sleep), whereas LLR shows up as "T" (stopped by job control). This difference makes no practical difference to me as a user, it was just unexpected.

Also ps lists the AP apps as being threaded, whereas LLR are not.

LLR is a standalone program which need a wrapper to be between it and Boinc. AP is a "native" (linked with Boinc libraries, could communicate with Boinc core directly). So some different approaches are used:
- When Boinc tell the LLR wrapper to pause, LLR has no way to pause calculations. So wrapper just freezes LLR process.
- When Boinc tell AP to pause, "native" program could indeed stop calculating and do only endless wait-for-events loop - check events state (was pause request cleared? was exit request received?); if no events, sleep for 1 second; check events again and so on. So you see it mostly in a sleep state.
- An extra thread in native app is automatically created by Boinc libraries to receive events from Boinc core. It runs small amount of code once per second or so, so you'll not see noticeable amount of CPU time spent in it.

Skivelitis2

Joined: 28 Mar 15
Posts: 42
ID: 388656
Credit: 283,325,222
RAC: 57,699

Message 100841 - Posted: 12 Nov 2016 | 1:41:32 UTC

Thanks for the clarification on the GPU apps, stream. I have 2 GT730 GPUs, a 2GB in a Win10 machine and a 1GB in a WinXP machine. I had hoped to upgrade one of the GPUs and "retire" the 1GB model so as to use both on AP27. Probably still will, I can keep one on GFN if I must. Just nice to have options but fully understand the whole XP thing. Maybe I'll see if I can come across a 64 bit copy of XP and do some testing. Used to have one somewhere.....

River~~

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

Message 100939 - Posted: 16 Nov 2016 | 7:42:32 UTC - in response to Message 100835.
Last modified: 16 Nov 2016 | 7:43:58 UTC

...
- When Boinc tell the LLR wrapper to pause, LLR has no way to pause calculations. So wrapper just freezes LLR process.
- When Boinc tell AP to pause, "native" program could indeed stop calculating and do only endless wait-for-events loop - check events state (was pause request cleared? was exit request received?); if no events, sleep for 1 second; check events again and so on. So you see it mostly in a sleep state.
- An extra thread in native app is automatically created by Boinc libraries to receive events from Boinc core. It runs small amount of code once per second or so, so you'll not see noticeable amount of CPU time spent in it.

Thanks for explaining,
River~~

[sorry if this is a double post, the connection reset during first attempt]

River~~

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

Message 100940 - Posted: 16 Nov 2016 | 7:56:03 UTC - in response to Message 100841.

Thanks for the clarification on the GPU apps, stream. I have 2 GT730 GPUs, a 2GB in a Win10 machine and a 1GB in a WinXP machine. I had hoped to upgrade one of the GPUs and "retire" the 1GB model so as to use both on AP27. Probably still will, I can keep one on GFN if I must. Just nice to have options but fully understand the whole XP thing. Maybe I'll see if I can come across a 64 bit copy of XP and do some testing. Used to have one somewhere.....

or maybe Linux? (Not sure if Linux has drivers for the card that's in the XP box.).

CPU tasks runs faster under Linux than Windows, they seem to natively run faster PLUS there is a MUCH smaller operating system overhead in my experience (ie the difference between "elapsed" and "cpu" times is about ten times as big under Windows. I have tested this for Win 7 and Win 10, each is slower for Boinc CPU than Linux. But of course that is no help if Linux wont run your Gpu...

R~~

Skivelitis2

Joined: 28 Mar 15
Posts: 42
ID: 388656
Credit: 283,325,222
RAC: 57,699

Message 100947 - Posted: 16 Nov 2016 | 12:26:37 UTC

Which distro would you recommend? As the machine is for all intents and purposes a dedicated cruncher, I would like the lightest weight version possible. Ease of setup and use would be a MAJOR consideration as well. My only experience with Linux was about 8 years ago. I ran it under a dual-boot program called Wubi I believe. Wanted to speed up output on the old Virtual Prairie project and remember the learning curve being rather steep for me at the time and that was CPU only.

Ross*

Joined: 22 Apr 10
Posts: 66
ID: 59181
Credit: 1,056,241,660
RAC: 0

Message 100987 - Posted: 16 Nov 2016 | 22:51:46 UTC

Hi
Can anyone tell why AMD cards are so hopeless at this AP27?
Ross*
____________

Van Zimmerman
Volunteer tester

Joined: 30 Aug 12
Posts: 1950
ID: 168418
Credit: 6,315,554,899
RAC: 2

Message 101007 - Posted: 17 Nov 2016 | 12:48:52 UTC - in response to Message 100987.

Hi
Can anyone tell why AMD cards are so hopeless at this AP27?
Ross*

I wish I could tell you why, but I feel your pain. They are absolutely terrible at AP, when they are perhaps not so terrible at other projects.

Using my (slightly overclocked) 1060 3g as a benchmark, approximate completion times for select projects are:
AP27 2280 seconds
GFN19 5616
PPS-Sieve 602

On a (slightly overclocked) Rx 480
AP27 6370 (+179%)
GFN19 6540 (+16%)
PPS-Sieve 647 (+7%)

On a R9 290X (running stock clocks and -50mv)
AP27 5828 (+156%)
GFN19 6780 (+21%)
PPS-Sieve 734 (+22%)

Those numbers kind of speak for themselves. Kepler gpus may not be immune from this either:

On a GTX 770 running stock under macOS
AP27 8300 (+264%)
PPS-Sieve 567 (-6%, average of 2 units running concurrently).

Keith
Volunteer tester

Joined: 8 Dec 13
Posts: 429
ID: 284516
Credit: 400,111,435
RAC: 115,634

Message 101010 - Posted: 17 Nov 2016 | 13:37:40 UTC - in response to Message 101007.

On a GTX 770 running stock under macOS
AP27 8300 (+264%)
PPS-Sieve 567 (-6%, average of 2 units running concurrently).

This makes me feel much better about the run times on my GTX760 for AP. I was thinking it was either being CPU bound or not getting enough power from the power supply. Now I know it's normal.
Thanks!
____________
My Primes
Badge Score: 2*1 + 4*2 + 6*5 + 7*8 + 9*2 + 10*1 = 124

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101019 - Posted: 17 Nov 2016 | 15:52:44 UTC

A new AP25 was just discovered: http://www.primegrid.com/forum_thread.php?id=7012&nowrap=true#101017

Will we find the first AP27 during the challenge?
____________
My lucky number is 75898524288+1

Rick Reynolds

Joined: 13 Aug 10
Posts: 7792
ID: 65483
Credit: 5,163,326,993
RAC: 766,911

Message 101023 - Posted: 17 Nov 2016 | 16:55:19 UTC - in response to Message 101019.

A new AP25 was just discovered: http://www.primegrid.com/forum_thread.php?id=7012&nowrap=true#101017

Will we find the first AP27 during the challenge?

Congrats on the find!!

Dj Ninja

Joined: 2 Jan 08
Posts: 23
ID: 17009
Credit: 385,365,738
RAC: 0

Message 101047 - Posted: 18 Nov 2016 | 14:07:38 UTC

As for GPU app, it was Michael's decision to not send it to 32-bit systems although GPU app itself is 32-bit, motivating that number of such systems (32-bit OS and capable GPU with 2+ GB of video RAM) was very low.

That's bad because I have such a system. GTX670 GPU on an older WinXP32 gaming rig I didn't want to rebuild yet. Because I can't archive full power due to your limitation not sending a 32 bit capable app to a 32 bit system I will *not* participate in the race.

Joe O

Joined: 13 Jul 07
Posts: 51
ID: 9819
Credit: 19,694,269
RAC: 0

Message 101050 - Posted: 18 Nov 2016 | 14:31:06 UTC

On a 32-bit system, each process is given 4 GB of virtual memory to play with, which is separated into 2 GB of user space that the application can actually use at a time.
Not only does 32-bit have a hard limit for the amount of memory it can address, there's also another problem: your devices, like your video card and motherboard BIOS take up room in that same 4 GB space, which means the underlying operating system gets access to even less of your RAM.
Windows expert Mark Russinovich found that a desktop running 32-bit Windows with 4 GB of RAM and two 1 GB video cards only had 2.2 GB of RAM available for the operating system—so the bigger and better your video cards get, the less of that 4 GB will be accessible on a 32-bit system.
____________
1437 · 2^495800 + 1

Volunteer moderator
Volunteer tester
Project scientist

Joined: 5 Feb 08
Posts: 1181
ID: 18646
Credit: 476,657,059
RAC: 113,974

Message 101062 - Posted: 18 Nov 2016 | 18:08:42 UTC

is there a noise out of the cooling systems in the ear/air ???
seems to me the challenge started ;)

let's have a party

\o/
____________
my current lucky number: 3749*2^1555697+1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/

Keith
Volunteer tester

Joined: 8 Dec 13
Posts: 429
ID: 284516
Credit: 400,111,435
RAC: 115,634

Message 101064 - Posted: 18 Nov 2016 | 18:29:35 UTC - in response to Message 101062.

is there a noise out of the cooling systems in the ear/air ???
seems to me the challenge started ;)

let's have a party

\o/

Supposedly anyway...
Can't check the leaderboard yet. *waits patiently*
____________
My Primes
Badge Score: 2*1 + 4*2 + 6*5 + 7*8 + 9*2 + 10*1 = 124

Mankka*

Joined: 2 Jun 10
Posts: 19
ID: 61697
Credit: 471,783,276
RAC: 0

Message 101066 - Posted: 18 Nov 2016 | 18:50:15 UTC - in response to Message 101064.

...I'm in no hurry with the stats as I run ATI/AMD GPUs ;)

Volunteer moderator
Volunteer tester
Project scientist

Joined: 5 Feb 08
Posts: 1181
ID: 18646
Credit: 476,657,059
RAC: 113,974

Message 101067 - Posted: 18 Nov 2016 | 18:52:54 UTC
Last modified: 18 Nov 2016 | 18:55:34 UTC

first of mine one should be at stats:

18.11.2016 19:42:56 Computation for task ap27_159055_0 finished

workunit <press button>
____________
my current lucky number: 3749*2^1555697+1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/

Ken_g6
Volunteer developer

Joined: 4 Jul 06
Posts: 903
ID: 3110
Credit: 176,622,625
RAC: 31,170

Message 101071 - Posted: 18 Nov 2016 | 19:16:17 UTC - in response to Message 101067.

Mine too. But stats appear empty right now.
____________

Williamd007

Joined: 14 Dec 10
Posts: 28
ID: 76891
Credit: 2,397,073,776
RAC: 12,098

Message 101072 - Posted: 18 Nov 2016 | 19:16:44 UTC

is it just me or have the stats for the wallis challenge stopped working?

Rick Reynolds

Joined: 13 Aug 10
Posts: 7792
ID: 65483
Credit: 5,163,326,993
RAC: 766,911

Message 101073 - Posted: 18 Nov 2016 | 19:19:15 UTC

must be just you.

no, kidding - don't think the stats are working yet

Williamd007

Joined: 14 Dec 10
Posts: 28
ID: 76891
Credit: 2,397,073,776
RAC: 12,098

Message 101074 - Posted: 18 Nov 2016 | 19:20:40 UTC - in response to Message 101073.

They were working earlier thou.

Volunteer moderator
Volunteer tester
Project scientist

Joined: 5 Feb 08
Posts: 1181
ID: 18646
Credit: 476,657,059
RAC: 113,974

Message 101075 - Posted: 18 Nov 2016 | 19:29:26 UTC

\o/
go, go, go ...
____________
my current lucky number: 3749*2^1555697+1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101077 - Posted: 18 Nov 2016 | 19:39:16 UTC

Stats are working now. I've also been opening up the flood gates so you can get more tasks than you could earlier. It's not quite up to the normal levels, but it should be back to normal within 15 minutes or so.
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101078 - Posted: 18 Nov 2016 | 19:42:51 UTC

The good news is Rafael found an AP25! Congratz! (More details later.)

The bad news is it was 37 minutes before the challenge started. :)
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101084 - Posted: 18 Nov 2016 | 21:30:03 UTC

There's now another AP25 that is a challenge task. Need to wait for the wingman before putting up that announcement. Let me go work on Rafael's...
____________
My lucky number is 75898524288+1

Crun-chi
Volunteer tester

Joined: 25 Nov 09
Posts: 2976
ID: 50683
Credit: 57,167,716
RAC: 56,558

Message 101086 - Posted: 18 Nov 2016 | 21:51:37 UTC - in response to Message 101084.

Two AP25 in hours: it looks like it will be productive challenge :)

____________
314187728^131072+1 GENERALIZED FERMAT :)
93*10^1029523-1 REPDIGIT PRIME
31*332^367560+1 CRUS PRIME
Proud member of team Aggie The Pew. Go Aggie!

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101087 - Posted: 18 Nov 2016 | 21:54:11 UTC

Congratulations Rafael on finding a new AP25!

322477370185894411+9633039*23#*n for n=0..24

Rafael's find wasn't part of the challenge, but the double check task was. :)
____________
My lucky number is 75898524288+1

Ken_g6
Volunteer developer

Joined: 4 Jul 06
Posts: 903
ID: 3110
Credit: 176,622,625
RAC: 31,170

Message 101091 - Posted: 19 Nov 2016 | 0:28:56 UTC

From a TeAm-mate of mine:

Sadly my 6950 does not have enough ram to run the program. I ran a test unit for primegrid just fine with only 1gb, but they limit the program to 1.5 for some reason.

Does AP27 really need 1.5GB VRAM?
____________

Crun-chi
Volunteer tester

Joined: 25 Nov 09
Posts: 2976
ID: 50683
Credit: 57,167,716
RAC: 56,558

Message 101093 - Posted: 19 Nov 2016 | 0:39:51 UTC - in response to Message 101091.

My 750Ti running AP26 unit

____________
314187728^131072+1 GENERALIZED FERMAT :)
93*10^1029523-1 REPDIGIT PRIME
31*332^367560+1 CRUS PRIME
Proud member of team Aggie The Pew. Go Aggie!

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101095 - Posted: 19 Nov 2016 | 3:02:05 UTC

We have another AP25! http://www.primegrid.com/forum_thread.php?id=7012&nowrap=true#101094

Congratulations Josef!

(There's two more AP25s in the pipeline.)
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101102 - Posted: 19 Nov 2016 | 14:48:44 UTC

We have two more AP25s!

Congratulations go to Ross for finding 84314670428700353+9954590*23#*n for n=0..24 and to John for finding 193513604089287343+10189009*23#*n for n=0..24.
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101117 - Posted: 19 Nov 2016 | 19:49:24 UTC

Status:

After a bit more than 1 day:

Challenge: Wallis is Born
App: 11 (AP27)
(As of 2016-11-19 19:40:31 UTC)

42222 tasks have been sent out. [CPU/GPU/anonymous_platform: 8728 (21%) / 33460 (79%) / 34 (0%)]

Of those tasks that have been sent out:

3540 (8%) came back with some kind of an error. [417 (1%) / 3123 (7%) / 0 (0%)]
21336 (51%) have returned a successful result. [1357 (3%) / 19953 (47%) / 26 (0%)]
17349 (41%) are still in progress. [6956 (16%) / 10385 (25%) / 8 (0%)]

Of the tasks that have been returned successfully:

8159 (38%) are pending validation. [414 (2%) / 7736 (36%) / 9 (0%)]
13177 (62%) have been successfully validated. [943 (4%) / 12217 (57%) / 17 (0%)]
0 (0%) were invalid. [0 (0%) / 0 (0%) / 0 (0%)]
0 (0%) are inconclusive. [0 (0%) / 0 (0%) / 0 (0%)]

What a great start to the challenge! We sent out about ten times the normal amount of tasks in the first day, which is fairly typical for challenges.

What isn't typical at all are the results. So far, we've found 5 AP25s since yesterday, including one that was found just before the challenge started (but the double check WAS part of the challenge, so I'm counting it), and another one that's still in the pipeline. To put things in perspective, since we started the AP27 search in September, a total of five AP25s (including one AP26) were found -- the same number as have been found in the last day. I can't really account for why the challenge has been so prolific at producing AP discoveries.
____________
My lucky number is 75898524288+1

mackerel
Volunteer tester

Joined: 2 Oct 08
Posts: 2420
ID: 29980
Credit: 416,309,981
RAC: 133,732

Message 101118 - Posted: 19 Nov 2016 | 20:12:16 UTC - in response to Message 101117.

I'm not up to speed to adequately understand the other post on what parts were or were not covered in the previous search. Could it be we're doing more of unsearched territory now?

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101120 - Posted: 19 Nov 2016 | 20:56:51 UTC - in response to Message 101118.

I'm not up to speed to adequately understand the other post on what parts were or were not covered in the previous search. Could it be we're doing more of unsearched territory now?

Slightly more, yes, but nowhere near enough to account for the sudden deluge of AP25 finds. When we started, we were doing 70% new work and 30 percent double checking. (6 out of 20 shifts were double checks). Right now we're doing 80% new work and 20% double checks (4 of 20). Furthermore, 4 of the 5 recent AP25s with in shift=640 tasks, and nothing at all has changed with those tasks.

It's probably just a statistical fluke.
____________
My lucky number is 75898524288+1

Crun-chi
Volunteer tester

Joined: 25 Nov 09
Posts: 2976
ID: 50683
Credit: 57,167,716
RAC: 56,558

Message 101122 - Posted: 20 Nov 2016 | 0:08:04 UTC - in response to Message 101117.

What a great start to the challenge! We sent out about ten times the normal amount of tasks in the first day, which is fairly typical for challenges.

I am surprised, that number of pending units raise, regardless 10 times more WU sent. I was expected faster validation time.

____________
314187728^131072+1 GENERALIZED FERMAT :)
93*10^1029523-1 REPDIGIT PRIME
31*332^367560+1 CRUS PRIME
Proud member of team Aggie The Pew. Go Aggie!

Keith
Volunteer tester

Joined: 8 Dec 13
Posts: 429
ID: 284516
Credit: 400,111,435
RAC: 115,634

Message 101125 - Posted: 20 Nov 2016 | 5:37:34 UTC - in response to Message 101122.

What a great start to the challenge! We sent out about ten times the normal amount of tasks in the first day, which is fairly typical for challenges.

I am surprised, that number of pending units raise, regardless 10 times more WU sent. I was expected faster validation time.

For the challenge people are running the AP tasks on hardware that they normally wouldn't use for AP. Some of that hardware being slower CPUs that take a long time to run units.
Normally I run a GTX 1060 (40 minutes a task) and a GTX 760 (2.5 hours a task). Now I also have an i7-4790 (16 hours a task) an Athlon X4 635 (48 hours a task) and a Pentium dual core E5500 (also 48 hours a task) running for the challenge. So you would be waiting a long time for tasks to validate with slow hardware like that running.
____________
My Primes
Badge Score: 2*1 + 4*2 + 6*5 + 7*8 + 9*2 + 10*1 = 124

joe carnivore

Joined: 22 May 09
Posts: 5
ID: 40459
Credit: 223,962,819
RAC: 75,641

Message 101126 - Posted: 20 Nov 2016 | 7:34:25 UTC - in response to Message 101125.

What a great start to the challenge! We sent out about ten times the normal amount of tasks in the first day, which is fairly typical for challenges.

I am surprised, that number of pending units raise, regardless 10 times more WU sent. I was expected faster validation time.

For the challenge people are running the AP tasks on hardware that they normally wouldn't use for AP. Some of that hardware being slower CPUs that take a long time to run units.
Normally I run a GTX 1060 (40 minutes a task) and a GTX 760 (2.5 hours a task). Now I also have an i7-4790 (16 hours a task) an Athlon X4 635 (48 hours a task) and a Pentium dual core E5500 (also 48 hours a task) running for the challenge. So you would be waiting a long time for tasks to validate with slow hardware like that running.

Thats right. my two Phenom x4 9450 are over 52%.

Dave

Joined: 13 Feb 12
Posts: 2779
ID: 130544
Credit: 931,529,915
RAC: 340,866

Message 101127 - Posted: 20 Nov 2016 | 8:21:24 UTC
Last modified: 20 Nov 2016 | 8:23:02 UTC

Celeron N2840 2 tasks 3.8 days :) (it normally does sieving).

Every little helps, as a certain unmentionable United Kingdom retail conglomerate keeps bleating.

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101137 - Posted: 20 Nov 2016 | 20:08:23 UTC

We've found another AP25!

203563934890169353+11105961*23#*n for n=0..24

Congratulations Stephen!

____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101138 - Posted: 20 Nov 2016 | 20:09:57 UTC

Here's the day 2 status update:

Challenge: Wallis is Born
App: 11 (AP27)
(As of 2016-11-20 20:08:30 UTC)

73274 tasks have been sent out. [CPU/GPU/anonymous_platform: 13905 (19%) / 59302 (81%) / 67 (0%)]

Of those tasks that have been sent out:

5970 (8%) came back with some kind of an error. [723 (1%) / 5247 (7%) / 0 (0%)]
47028 (64%) have returned a successful result. [5003 (7%) / 41978 (57%) / 47 (0%)]
20276 (28%) are still in progress. [8179 (11%) / 12077 (16%) / 20 (0%)]

Of the tasks that have been returned successfully:

12346 (26%) are pending validation. [936 (2%) / 11397 (24%) / 13 (0%)]
34679 (74%) have been successfully validated. [4067 (9%) / 30578 (65%) / 34 (0%)]
3 (0%) were invalid. [0 (0%) / 3 (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: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101174 - Posted: 22 Nov 2016 | 14:07:35 UTC

There's a bit more than one day to go!

Challenge: Wallis is Born
App: 11 (AP27)
(As of 2016-11-22 14:02:54 UTC)

123694 tasks have been sent out. [CPU/GPU/anonymous_platform: 22325 (18%) / 101246 (82%) / 123 (0%)]

Of those tasks that have been sent out:

9491 (8%) came back with some kind of an error. [1696 (1%) / 7795 (6%) / 0 (0%)]
92478 (75%) have returned a successful result. [11763 (10%) / 80616 (65%) / 99 (0%)]
21725 (18%) are still in progress. [8866 (7%) / 12835 (10%) / 24 (0%)]

Of the tasks that have been returned successfully:

15270 (17%) are pending validation. [1364 (1%) / 13884 (15%) / 22 (0%)]
77198 (83%) have been successfully validated. [10399 (11%) / 66722 (72%) / 77 (0%)]
6 (0%) were invalid. [0 (0%) / 6 (0%) / 0 (0%)]
4 (0%) are inconclusive. [0 (0%) / 4 (0%) / 0 (0%)]

So far we've found 5 AP25s plus 3 more AP25s that are waiting for wingmen. With 28 hours to go, there's still time to find an AP26 or even the AP27 we're searching for!
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101196 - Posted: 23 Nov 2016 | 2:51:33 UTC

With about 15 hours to go, it's time for the usual end of challenge reminder:

At the Conclusion of the Challenge

After the challenge ends, we would prefer users "moving on" to finish those tasks they have downloaded, if not then please ABORT the WU's instead of DETACHING, RESETTING, or PAUSING.

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

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101199 - Posted: 23 Nov 2016 | 5:19:51 UTC
Last modified: 23 Nov 2016 | 5:20:09 UTC

Another AP25!

231349139005158193+12169920*23#*n for n=0..24

Congratulations Stefan!

There's at least two more still in the pipeline.
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101219 - Posted: 23 Nov 2016 | 18:18:07 UTC

The challenge is over! Great job everyone.

Final statistics:

Challenge: Wallis is Born
App: 11 (AP27)
(As of 2016-11-23 18:09:32 UTC)

154346 tasks have been sent out. [CPU/GPU/anonymous_platform: 26133 (17%) / 128088 (83%) / 125 (0%)]

Of those tasks that have been sent out:

16461 (11%) came back with some kind of an error. [4196 (3%) / 12265 (8%) / 0 (0%)]
123301 (80%) have returned a successful result. [16733 (11%) / 106461 (69%) / 107 (0%)]
14395 (9%) are still in progress. [5163 (3%) / 9214 (6%) / 18 (0%)]

Of the tasks that have been returned successfully:

13078 (11%) are pending validation. [1268 (1%) / 11802 (10%) / 8 (0%)]
110204 (89%) have been successfully validated. [15465 (13%) / 94640 (77%) / 99 (0%)]
15 (0%) were invalid. [0 (0%) / 15 (0%) / 0 (0%)]
4 (0%) are inconclusive. [0 (0%) / 4 (0%) / 0 (0%)]

We found a total of 8 AP25 during the challenge, including two that I haven't announced yet because I'm waiting for the wingmen to finish.
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101220 - Posted: 23 Nov 2016 | 18:19:29 UTC

Now the cleanup starts. I expect it to take between 3 and 6 weeks.

Cleanup Status:
23-Nov: Wallis is Born: 13054 tasks outstanding; 11063 affecting individual (292) scoring positions; 5750 affecting team (63) scoring positions.
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101237 - Posted: 23 Nov 2016 | 21:44:37 UTC

Here's one of the remaining AP25 discoveries:

290884103695102903+12444168*23#*n for n=0..24

Congratulations Mike!

(There's one more to announce.)
____________
My lucky number is 75898524288+1

Gary Craig
Volunteer tester

Joined: 30 Dec 09
Posts: 3213
ID: 52890
Credit: 1,005,618,748
RAC: 0

Message 101241 - Posted: 23 Nov 2016 | 23:32:06 UTC
Last modified: 23 Nov 2016 | 23:36:29 UTC

When there are ties in the challenge score between individuals, is the "rank" determined by reporting time (of day)? Or is the sorting random? Something else? Clearly not alphabetical by name or team.

Thanks all for a good challenge.

--Gary

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101248 - Posted: 24 Nov 2016 | 2:29:02 UTC - in response to Message 101241.

When there are ties in the challenge score between individuals, is the "rank" determined by reporting time (of day)? Or is the sorting random? Something else? Clearly not alphabetical by name or team.

Thanks all for a good challenge.

--Gary

For challenges like this one, where all the tasks have the exact same value, it's a complete tie in the rankings. If, for example, positions 2, 3, and 4 have the same score, then they are tied for 2nd; the order in which they are listed is random.

For most challenges, where each task has a unique score, even if two people appear to be tied (i.e., the score are equal as far as the displayed digits show), there will always be SOME difference between two scores, even if you can't see it. The best example would be an SGS challenge where the differences between the tasks are 5 or 6 digits to the right of the decimal point. For most challenges, there's no ties.

But for an AP27 challenge, literal ties will happen.
____________
My lucky number is 75898524288+1

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101251 - Posted: 24 Nov 2016 | 3:11:42 UTC

The final AP25 can now be revealed:

183800317923336901+10816541*23#*n for n=0..24

Congratulations James!
____________
My lucky number is 75898524288+1

Jay

Joined: 27 Feb 10
Posts: 99
ID: 56067
Credit: 53,145,442
RAC: 10,757

Message 101253 - Posted: 24 Nov 2016 | 4:51:13 UTC - in response to Message 101251.

The final AP25 can now be revealed:

183800317923336901+10816541*23#*n for n=0..24

Congratulations James!

Yes, congratulations to James and everyone else who found an AP25.

I'm not really familiar with this project so my questions might be stupid, but from what I read an AP has to be more than a single digit. I'm guessing an AP2, or AP3 are quite common. Is there anyway to give a probability for AP's of certain lengths in the area we are looking? For example, would 95% of the tests find an AP2, and 50% find an AP8 and 1% find an AP20, etc? Where is the break where a test goes from more than 50% likely to less than 50% likely?

Thanks,
Jay

mackerel
Volunteer tester

Joined: 2 Oct 08
Posts: 2420
ID: 29980
Credit: 416,309,981
RAC: 133,732

Message 101258 - Posted: 24 Nov 2016 | 11:04:23 UTC - in response to Message 101253.

Not an expert here, but AP2 would just be a pair of primes so you could have fun working out how big an infinity there are of those compared to primes themselves (infinity factorial maybe?).

So in that sense, only AP3 and above would be worth considering.

I don't know where I saw it, but I vaguely recall there is an expected ratio of how frequent different AP lengths are. I think for every extra length it was a factor of 4, but my memory could well be wrong there.

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101262 - Posted: 24 Nov 2016 | 11:34:14 UTC - in response to Message 101220.

Cleanup Status:
23-Nov: Wallis is Born: 13054 tasks outstanding; 11063 affecting individual (292) scoring positions; 5750 affecting team (63) scoring positions.
24-Nov: Wallis is Born: 7074 tasks outstanding; 5250 affecting individual (274) scoring positions; 930 affecting team (38) scoring positions.

____________
My lucky number is 75898524288+1

Grebuloner
Volunteer tester

Joined: 2 Nov 09
Posts: 348
ID: 49572
Credit: 1,673,289,334
RAC: 1,601,972

Message 101281 - Posted: 24 Nov 2016 | 17:41:26 UTC - in response to Message 101251.

Hot damn!

And I see it wasn't on a computer beset by the ratties.

On a related note, though, I didn't get an email or PM in notification. Are these not turned on for the significant finds? If not, can they be, please?
____________
Eating more cheese on Thursdays.

JeppeSN

Joined: 5 Apr 14
Posts: 1266
ID: 306875
Credit: 17,694,119
RAC: 35,588

Message 101288 - Posted: 24 Nov 2016 | 22:46:04 UTC - in response to Message 101258.

Not an expert here, but AP2 would just be a pair of primes

Absolutely, so Jens Kruse Andersen's AP record page does not have them.

Two points are always on a straight line, it takes at least three points to make collinearity something interesting.

so you could have fun working out how big an infinity there are of those compared to primes themselves (infinity factorial maybe?).

I am thinking "infinity choose two", that is (infinity)*(infinity - 1) / 2.

/JeppeSN

mackerel
Volunteer tester

Joined: 2 Oct 08
Posts: 2420
ID: 29980
Credit: 416,309,981
RAC: 133,732

Message 101289 - Posted: 24 Nov 2016 | 23:24:01 UTC - in response to Message 101288.

so you could have fun working out how big an infinity there are of those compared to primes themselves (infinity factorial maybe?).

I am thinking "infinity choose two", that is (infinity)*(infinity - 1) / 2.

/JeppeSN

That'll be the one. I got confused between what I know as the handshake problem: how many ways can n people shake hands with everyone else, and the number of possible ways of arranging n items. This is why I leave maths to people better at it!

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101298 - Posted: 25 Nov 2016 | 11:09:51 UTC - in response to Message 101262.

Cleanup Status:

23-Nov: Wallis is Born: 13054 tasks outstanding; 11063 affecting individual (292) scoring positions; 5750 affecting team (63) scoring positions.
24-Nov: Wallis is Born: 7074 tasks outstanding; 5250 affecting individual (274) scoring positions; 930 affecting team (38) scoring positions.
25-Nov: Wallis is Born: 4668 tasks outstanding; 3154 affecting individual (252) scoring positions; 440 affecting team (28) scoring positions.

____________
My lucky number is 75898524288+1

River~~

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

Message 101300 - Posted: 25 Nov 2016 | 12:31:59 UTC - in response to Message 101220.

...
Cleanup Status:
23-Nov: Wallis is Born: 13054 tasks outstanding; 11063 affecting individual (292) scoring positions; 5750 affecting team (63) scoring positions.

I an guessing this might be why the Wallis positions have not yet been included in the overall position for the year so far? As it is only the second challenge Ive participated in this year, I hope to be going up rather than down once my Wallis score joins Pi-Paddy.
(http://www.charleygielkens.nl/pg/index.php?challenge=0&uot=u&year=2016&team=&type=html)

R~~

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101303 - Posted: 25 Nov 2016 | 15:08:05 UTC - in response to Message 101300.

...
Cleanup Status:
23-Nov: Wallis is Born: 13054 tasks outstanding; 11063 affecting individual (292) scoring positions; 5750 affecting team (63) scoring positions.

I an guessing this might be why the Wallis positions have not yet been included in the overall position for the year so far? As it is only the second challenge Ive participated in this year, I hope to be going up rather than down once my Wallis score joins Pi-Paddy.
(http://www.charleygielkens.nl/pg/index.php?challenge=0&uot=u&year=2016&team=&type=html)

R~~

Good guess, but only partly correct.

The individual standings are not updated until the challenges are complete, so that's indeed one reason why they are not updated yet. The more important reason is that the person who does that has been away, so it hasn't been updated in a few months. There's no ETA on when it will get updated.
____________
My lucky number is 75898524288+1

Mankka*

Joined: 2 Jun 10
Posts: 19
ID: 61697
Credit: 471,783,276
RAC: 0

Message 101304 - Posted: 25 Nov 2016 | 15:44:18 UTC - in response to Message 101281.
Last modified: 25 Nov 2016 | 16:01:55 UTC

Hot damn!

And I see it wasn't on a computer beset by the ratties.

On a related note, though, I didn't get an email or PM in notification. Are these not turned on for the significant finds? If not, can they be, please?

@Grebuloner: New AP finds are announced in this thread =)

(and they are also listed on your account page under AP26 tasks)

Mankka*

Grebuloner
Volunteer tester

Joined: 2 Nov 09
Posts: 348
ID: 49572
Credit: 1,673,289,334
RAC: 1,601,972

Message 101311 - Posted: 26 Nov 2016 | 1:53:13 UTC - in response to Message 101304.

Hot damn!

And I see it wasn't on a computer beset by the ratties.

On a related note, though, I didn't get an email or PM in notification. Are these not turned on for the significant finds? If not, can they be, please?

@Grebuloner: New AP finds are announced in this thread =)

(and they are also listed on your account page under AP26 tasks)

Mankka*

True, but even non top 5000 genefer 15s get you an email/PM. It's nice to be at work, or wake up and see an email from PG that you found something special, and aren't AP25+'s special?
____________
Eating more cheese on Thursdays.

Mankka*

Joined: 2 Jun 10
Posts: 19
ID: 61697
Credit: 471,783,276
RAC: 0

Message 101319 - Posted: 26 Nov 2016 | 13:26:29 UTC - in response to Message 101311.
Last modified: 26 Nov 2016 | 13:28:08 UTC

True, but even non top 5000 genefer 15s get you an email/PM. It's nice to be at work, or wake up and see an email from PG that you found something special, and aren't AP25+'s special?

I totally agree with you and that's why I'm banging my head in the wall with my ATI/AMDs even if their performance sucks badly in this one (they shouldn't, but I guess it's about how the apps are written), to find the first AP27 =)

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 101320 - Posted: 26 Nov 2016 | 13:49:14 UTC - in response to Message 101298.
Last modified: 13 Dec 2016 | 14:00:10 UTC

Cleanup Status:

23-Nov: Wallis is Born: 13054 tasks outstanding; 11063 affecting individual (292) scoring positions; 5750 affecting team (63) scoring positions.
24-Nov: Wallis is Born: 7074 tasks outstanding; 5250 affecting individual (274) scoring positions; 930 affecting team (38) scoring positions.
25-Nov: Wallis is Born: 4668 tasks outstanding; 3154 affecting individual (252) scoring positions; 440 affecting team (28) scoring positions.
26-Nov: Wallis is Born: 3661 tasks outstanding; 2105 affecting individual (237) scoring positions; 244 affecting team (26) scoring positions.
27-Nov: Wallis is Born: 2856 tasks outstanding; 1450 affecting individual (214) scoring positions; 114 affecting team (20) scoring positions.
28-Nov: Wallis is Born: 2172 tasks outstanding; 1007 affecting individual (190) scoring positions; 57 affecting team (13) scoring positions.
29-Nov: Wallis is Born: 1715 tasks outstanding; 757 affecting individual (164) scoring positions; 46 affecting team (9) scoring positions.
30-Nov: Wallis is Born: 1311 tasks outstanding; 531 affecting individual (142) scoring positions; 36 affecting team (7) scoring positions.
1-Dec: Wallis is Born: 672 tasks outstanding; 214 affecting individual (90) scoring positions; 17 affecting team (5) scoring positions.
2-Dec: Wallis is Born: 240 tasks outstanding; 70 affecting individual (47) scoring positions; 3 affecting team (2) scoring positions.
3-Dec: Wallis is Born: 138 tasks outstanding; 36 affecting individual (29) scoring positions; 1 affecting team (1) scoring positions.
4-Dec: Wallis is Born: 94 tasks outstanding; 26 affecting individual (22) scoring positions; 1 affecting team (1) scoring positions.
5-Dec: Wallis is Born: 62 tasks outstanding; 19 affecting individual (17) scoring positions; 0 affecting team (0) scoring positions.
6-Dec: Wallis is Born: 51 tasks outstanding; 16 affecting individual (14) scoring positions; 0 affecting team (0) scoring positions.
7-Dec: Wallis is Born: 26 tasks outstanding; 5 affecting individual (5) scoring positions; 0 affecting team (0) scoring positions.
8-Dec: Wallis is Born: 23 tasks outstanding; 3 affecting individual (3) scoring positions; 0 affecting team (0) scoring positions.
9-Dec: Wallis is Born: 17 tasks outstanding; 2 affecting individual (2) scoring positions; 0 affecting team (0) scoring positions.
10-Dec: Wallis is Born: 15 tasks outstanding; 2 affecting individual (2) scoring positions; 0 affecting team (0) scoring positions.
11-Dec: Wallis is Born: 13 tasks outstanding; 2 affecting individual (2) scoring positions; 0 affecting team (0) scoring positions.
12-Dec: Wallis is Born: 9 tasks outstanding; 1 affecting individual (1) scoring positions; 0 affecting team (0) scoring positions.
13-Dec: Wallis is Born: 8 tasks outstanding; 1 affecting individual (1) scoring positions; 0 affecting team (0) scoring positions.
____________
My lucky number is 75898524288+1

Roger
Volunteer developer
Volunteer tester

Joined: 27 Nov 11
Posts: 1137
ID: 120786
Credit: 266,956,799
RAC: 7,867

Message 102193 - Posted: 14 Dec 2016 | 21:38:15 UTC

The results are final!

Top 3 individuals:

1: zunewantan
2: tng*
3: RaymondFO*

Top 3 teams:

2: Crunching@EVGA
3: SETI.Germany

Congratulations to the winners, and well done to everyone who participated.
See you at the next challenge!
____________

DeleteNull
Volunteer tester

Joined: 6 Apr 06
Posts: 238
ID: 2663
Credit: 6,812,247,428
RAC: 5,968,922

Message 102211 - Posted: 15 Dec 2016 | 8:16:37 UTC

As a first test:

The user stats 2016
____________
DeleteNull

TheDawgz

Joined: 7 Jul 08
Posts: 2235
ID: 25193
Credit: 568,161,626
RAC: 127,573

Message 102230 - Posted: 15 Dec 2016 | 15:43:09 UTC - in response to Message 102211.

As a first test:

The user stats 2016

Looks great! Can't wait for the Team stats!!

Danke sehr!
____________
There's someone in our head but it's not us.

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 102231 - Posted: 15 Dec 2016 | 16:19:51 UTC - in response to Message 102211.

As a first test:

The user stats 2016

Very nice.

You might (or might not) be aware of one little detail: Sometimes ties are ties, and sometimes they're not.

With most challenges, there's no possibility of a tie. Even if the numbers look identical in the standings, there will be differences further to the right of the decimal point. This is because no two tasks get the exact same credit. Even SGS tasks vary VERY slightly in credit.

With challenges where all the tasks do have the exact same credit (most sieves and AP27), however, a tie in the standings really is a tie, so you need to account for that in the point distribution. If 3rd and 4th place have identical scores, they should get the same number of points.

I said "most" sieves because GCW sieve will be a little different. Tasks in each base have identical credit, but each base has slightly different credit than any other base. An apparent tie in the standings might, or might not, be an actual tie. I'll probably need to revise the challenge software to explicitly show which rankings are real ties. It wouldn't hurt to remind me about this when we get to the GCW Sieve challenge, in case I forget.
____________
My lucky number is 75898524288+1

DeleteNull
Volunteer tester

Joined: 6 Apr 06
Posts: 238
ID: 2663
Credit: 6,812,247,428
RAC: 5,968,922

Message 102246 - Posted: 15 Dec 2016 | 23:03:37 UTC - in response to Message 102231.

To explain how the calculation is implemented:

I wrote down a few lines of Java code.....
The code reads the html of each challenge and converts it into raw data for the calculation of the points.
So the (user) score is a decimal number with only 2 digits, and......
in the moment the software assigns the points by the place a user has; place 3 gets more points than place 4.

I have to add an attribute to each challenge:
tie=true (tie is tie), or tie=false (points by place, because a tie isn't a tie)
The rule for a tie is? ("points_place_x" + "points_place_x+1")/2 for both users?!

The challenges with tie=true are:
Challenge_5 (TRP-Sieve), Challenge_8 (AP27) and the upcoming Challenge_9.

Is this o.k.?

____________
DeleteNull

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 102247 - Posted: 16 Dec 2016 | 1:46:02 UTC - in response to Message 102246.

To explain how the calculation is implemented:

I wrote down a few lines of Java code.....
The code reads the html of each challenge and converts it into raw data for the calculation of the points.
So the (user) score is a decimal number with only 2 digits, and......
in the moment the software assigns the points by the place a user has; place 3 gets more points than place 4.

I have to add an attribute to each challenge:
tie=true (tie is tie), or tie=false (points by place, because a tie isn't a tie)
The rule for a tie is? ("points_place_x" + "points_place_x+1")/2 for both users?!

The challenges with tie=true are:
Challenge_5 (TRP-Sieve), Challenge_8 (AP27) and the upcoming Challenge_9.

Is this o.k.?

Almost...

The rule for a tie is? ("points_place_x" + "points_place_x+1")/2 for both users?!

I was wondering about that myself. :)

I'm not sure how Charley did it, but usually when there's a tie, both people get the points for the better spot. So if A and B tie for third place, they both get the full points for 3rd (not the average of 3rd and 4th).

Theoretically speaking, if 300 people all tied for 1st place, they'd all get the full points for 1st. :)

____________
My lucky number is 75898524288+1

DeleteNull
Volunteer tester

Joined: 6 Apr 06
Posts: 238
ID: 2663
Credit: 6,812,247,428
RAC: 5,968,922

Message 102285 - Posted: 16 Dec 2016 | 15:45:32 UTC - in response to Message 102247.

.....
I'm not sure how Charley did it, but usually when there's a tie, both people get the points for the better spot. So if A and B tie for third place, they both get the full points for 3rd (not the average of 3rd and 4th).

Theoretically speaking, if 300 people all tied for 1st place, they'd all get the full points for 1st. :)

Here are the new stats for this rule (new URL).

Several users have a few points more. The first change in the ranking is on place 72.
____________
DeleteNull

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13385
ID: 53948
Credit: 227,918,956
RAC: 157,517

Message 102393 - Posted: 18 Dec 2016 | 14:58:29 UTC - in response to Message 102285.

.....
I'm not sure how Charley did it, but usually when there's a tie, both people get the points for the better spot. So if A and B tie for third place, they both get the full points for 3rd (not the average of 3rd and 4th).

Theoretically speaking, if 300 people all tied for 1st place, they'd all get the full points for 1st. :)

Here are the new stats for this rule (new URL).

Several users have a few points more. The first change in the ranking is on place 72.

As of right now, the challenge stats handle ties more intelligently. If two people are tied for third place, they'll now BOTH show "3" as their position. You no longer will need to guess whether two people who appear to have the same score are actually tied or not.

This only affects challenges starting with the current Winter Solstice challenge. Stats for earlier challenges can't be changed. It's not possible to regenerate those pages.
____________
My lucky number is 75898524288+1

DeleteNull
Volunteer tester

Joined: 6 Apr 06
Posts: 238
ID: 2663
Credit: 6,812,247,428
RAC: 5,968,922

Message 102432 - Posted: 19 Dec 2016 | 8:06:51 UTC

Thanks for the update. This will make the calculation of the stats easier.
Perhaps we need another thread to put in the information....

Current (user) stats are here, team stats will follow in about two days.
____________
DeleteNull

Message boards : Number crunching : Wallis is Born Challenge