PrimeGrid Winter Solstice Challenge (part 9 of the 2016 PrimeGrid Challenge Series)

Gilthanis

Gilthanis

[H]ard|DCer of the Year - 2014
Joined
Jan 29, 2006
Messages
8,677
Proth Prime Search (Sieve)

http://www.primegrid.com/forum_thread.php?id=7138#101309
December 18th 10:44:00 UTC until December 21st 10:44:00 UTC we will be having a 3 day challenge

This is part 9 of 9 challenges. It is CPU and GPU. Our team is currently in 12th place for the series.

Current team standings are: There is still cleanup from previous challenges so these positions are not final
25 - RFGuy_KCCO
101 - Coleslaw
105 - pututu
142 - HighTech67
268 - tjmagneto
285 - brilong
392 - travanx

Tips:
You must log into your account at PrimeGrid to change your project preferences so that you are only crunching the correct sub project. In this case PPS Sieve.
Being a GPU project, you may not want to run CPU work as GPU's will be much faster.
Intel boxes tend to perform better. The HT probably needs turned off since these are not sieve work units. Do your testing in advance.
Intel chips tend to out perform AMD chips
PrimeGrid work will stress hardware harder and so is more sensitive to overclocking. Therefore make sure you have tested their work units on your rigs before the challenge to make sure they are stable.
These challenges only count work units downloaded during and completed before end of the challenge. So, no bunkering.
 
I'm in. Anyone else joining in? Aiming to get turquoise badge (5M) for PPS Sieve.
 
Trying to get a couple M40's online. A lot of my formerly idle hardware is no longer idle, so that will cut into my usual output.
 
So 6:42 and 6:48 for the first pair of WUs.

Two unfortunate items:

  1. These are passively cooled cards, so am a little concerned about heat. Just rebooted and cranked up the fans in this server as high as the will go. Ideally I should move one of these cards to a different box, but I had to come up with a creative solution for powering these damned things, so two systems would just be that much bigger of a PITFA.

  2. The PCIe slots are only 8x electrical and I'm not sure how badly that hurts things vs. true 16x slots. Is a dual E5 Haswell box at least.
*edit* GPUs are hanging out at 69-72*C. Slow down limit is 89* and shut down is 92*, so figure it'll be fine. Just noticed the CPU ETA @ ~1D 6H. Ouch. Just set the BOINC processor number to match real cores, will see how much that helps before tinkering with it more.
 
Last edited:
fastgeek said:
Trying to get a couple M40's online. A lot of my formerly idle hardware is no longer idle, so that will cut into my usual output.
Click to expand...
"a couple" means like 10? Each tesla M40 costs around $4K each. Might as well get ten of 1070 hah hah:p

In the last PG challenge AP27, my i7-2600K at 4.2GHz completed one WU in 21.3 hours. At least you can do two rounds.:)
upload_2016-12-16_22-17-37.png
 
About 17 hours to go before the challenge starts. That is Sunday, Dec 18 at 2:44am PST or 5:44am EST. I will crunch before I go to bed.
 
Nope, just two. Borrowed them for a couple weeks from a different project. Tried to get more, but unfortunately there weren't any idle cards available. Didn't fully realize when this challenge started, thought it was Monday PST, so might not be bringing much more than these cards to the fight.
 
After 1 day and 5 hours of folding, we are at 20th position.

Team members contribution:
Rank Name Score
35 RFGuy_KCCO 4412639
80 fastgeek 1840566
129 pututu 1061865
417 HighTech67 202260

Did try CPU crunching on E5-2695v2 at 3GHz (10 cores), WU time is about 14.4 hours. Certainly not efficient at all.
 
I should have at least one box doing a little bit of work shortly. Unfortunately GPU crunching is light on my part these days.
 
Next year proposed PG challenge will have 90% (54/60 days) of CPU crunching. We seem to have good turn out in WCG and FAH projects. The rest, kind of .......:whistle:
 
Well, the majority of the team favors having a single project or one project for CPU and one for GPU depending on what is going on at that time. Only a small hand full of us like the variety that BOINC brings and embrace it. As more get familiar with BOINC, that may change, but it is a very long battle to get the team to switch to a new way of things.
 
Let's keep sharing and publishing BOINC projects since this is all about distributed computing. I realized that we lack new blood in the team. I think I'm probably the newest active member here :D It's almost one year anniversary since I joined [H] team (Dec 27, 2015). Just my one cent observation. Wow how time has passed by...
 
Yup, right now GPUs just aren't my strong suit. Maybe that'll be different in the future; but for now it's all about CPUs. That being said, nice to see you in the mix, RFGuy_KCCO! :D
 
It appears that this last PG challenge of the year is going to be my worst performance in terms of ranking :( A quick glance at those above me, most have upgraded to Pascal while I'm still on Maxwell (one 970 and one 960). Since I'm volunteering for science and to some extent be somewhat efficient at doing this, I'm contend as it is. Looking ahead Volta is coming next year and another round of upgrade for these folks, I presume. Good for them if they have spare cash.
 
Last edited:
We finished at 21st place. Thanks for participating.

Final scoring positions and credits:
Rank Name Score
38 RFGuy_KCCO 11275995
85 fastgeek 4604786
152 pututu 2605783
422 HighTech67 586554
717 Jathanis 121356

The next PG challenge in 2017 is scheduled to begin on Jan 3, 2017 and it is CPU only WUs.
 
You must log in or register to reply here.
Back
Top