121 - Happy St Patrick's Equinox Challenge - 16-21 March

Announced is a challenge at the 121 prime search project from the 16th till 21st of March. I suggest we start at 12:00 UTC and end at 12:00 UTC.

For this Challenge we're celebrating Happiness:
Lyrics: (Because I'm happy) / Clap along if you feel like a room without a roof / (Because I'm happy) / Clap along if you feel like happiness is the truth…
The International Day of Happiness is celebrated throughout the world on the 20th of March!

and St Patrick's Day:
St. Patrick’s Day occurs annually on March 17 in observance of the death of St. Patrick, the patron saint of Ireland. What began as a religious feast day in the 17th century has evolved into a variety of festivals across the globe celebrating Irish culture with parades, special foods, music, dancing, and a whole lot of green.

and the March equinox:
March 20 - March Equinox. The March equinox occurs at 10:29 UTC. The Sun will shine directly on the equator and there will be nearly equal amounts of day and night throughout the world. This is also the first day of spring (vernal equinox) in the Northern Hemisphere and the first day of fall (autumnal equinox) in the Southern Hemisphere.

More about the 27121 prime search can be found here. News and infos about the PRPNet client can be found here.

To take part, you have to activate the following lines in prpclient.ini:

bugfix:
1) link in stats page points now to this thread
2) sum line only for users, because users may not in any team (prevent confusion)
____________
Sysadm@Nbg
my current lucky number: 3651*2^1521717+1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/

Less than 30 minutes until the Challenge start.
I am heading to Rosie O'Grady's tomorrow night for St Patricks Day. So great being on a Friday night.
http://www.rosieogradys.com.au/

Due to celebrations on my own behalf I missed the start.
But everything seems to be going well ...
Happy racing!
____________
Sysadm@Nbg
my current lucky number: 3651*2^1521717+1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/

D:\PrimeGrid\prpclient\prpclient-1>prpclient.exe [2017-03-16 13:05:09 GST] PRPNet Client application v5.4.0 started [2017-03-16 13:05:09 GST] User name TroubledBunny at email address is [2017-03-16 13:05:10 GST] 121: Getting work from server prpnet.primegrid.com at port 12001 [2017-03-16 13:05:11 GST] 121: PRPNet server is version 5.4.0 k=121 Prime Search Starting Lucas Lehmer Riesel prime test of 121*2^7XXXXXX-1 Using AVX FFT length 448K, Pass1=448, Pass2=1K V1 = 4 ; Computing U0...done. Iter: 1725651/7XXXXXX, ERROR: ROUND OFF (0.4375) > 0.4 Continuing from last save file. Resuming LLR test of 121*2^XXXXXX-1 at iteration 1542273 [20.92%] Disregard last error. Result is reproducible and thus not a hardware problem. For added safety, redoing iteration using a slower, more reliable method. Continuing from last save file. Resuming LLR test of 121*2^XXXXXX-1 at iteration 1725651 [23.41%] Iter: 4782712/7XXXXXX, ERROR: ROUND OFF (0.4375) > 0.4 Continuing from last save file. Resuming LLR test of 121*2^XXXXXX-1 at iteration 4664193 [63.29%] Disregard last error. Result is reproducible and thus not a hardware problem. For added safety, redoing iteration using a slower, more reliable method. Continuing from last save file. Resuming LLR test of 121*2^XXXXXX-1 at iteration 4782712 [64.90%] Iter: 5193425/7XXXXXX, ERROR: ROUND OFF (0.40625) > 0.4 Continuing from last save file. Resuming LLR test of 121*2^XXXXXX-1 at iteration 4782713 [64.90%] Disregard last error. Result is reproducible and thus not a hardware problem. For added safety, redoing iteration using a slower, more reliable method. Continuing from last save file. Resuming LLR test of 121*2^XXXXXX-1 at iteration 5193425 [70.47%] 121*2^XXXXXX-1, iteration : 5300000 / XXXXXX [71.92%]. Time per iteration : 3.047 ms.

What do you think? Should I buy a new cpu cooler? It does gurgle a lot :)

I think that if you're not sure if your computer is reliable that you should absolutely, positively, not be running PRPNet. Run BOINC tasks instead so that your results can be verified.

You're near the maximum size candidate that FFT size can handle. If you ran it with -oFFT_Increment=1 then I don't think there would be any errors at all. Our BOINC validator ignores an error when it's followed by the "Disregard this error" line. In fact the program might switch to the next larger FFT all by itself after six or so "errors".

For a bit of background, the + side of PRPNet 121 and 27 projects are double checked on BOINC, although it is far behind. Currently:
k=121 up to n=3,049,908 on BOINC, n=7,388,708 on PRPNet (+ side)
k=27 up to n=3,321,956 on BOINC, n=6,427,079 on PRPNet (+ side)
121 and 27 candidates are regularly uploaded from the latest PPS Sieve files for both the plus and minus sides.
PPS Sieve is doing k's 5-9999 on both the plus and minus sides. k=3 is sieved separately.

The Challenge has ended. Congratulation to Grebuloner and The_Knights_Who_Say_Ni!

The top of the challenge rankings is as follows:

permalink for the stats of this challenge: http://u-g-f.de/PRPNet/2017_03_121_Challenge.html
____________
Sysadm@Nbg
my current lucky number: 3651*2^1521717+1
PSA-PRPNet-Stats-URL: http://u-g-f.de/PRPNet/