PrimeGrid Winter Solstice Challenge December 18-21, with GPUs!

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,284
3,905
75
Welcome to the PrimeGrid Winter Solstice Challenge. We're rounding out another year with a sieve challenge; but this is no ordinary sieve challenge. This challenge lets you sieve with GPUs! This challenge also happens to use a sieve client I wrote, so I have a few more suggestions than usual. :sneaky:

What you need to participate:
  • One or more fast x86 processors. (Even slow ones will do!)
  • Windows, Linux, or MacOS 10.4+.
  • BOINC, attached to PrimeGrid (http://www.primegrid.com/).
  • Your PrimeGrid Preferences with only "Proth Prime Search (Sieve)" selected in the Projects section. (It's the one marked as this month's Challenge project.)

What would help a whole lot:
  • For all your WUs to be downloaded after December 18, at 23:03 UTC.
  • An Nvidia GPU, of the 8000 series or later (so it's CUDA capable). A mid-range GTX 460 does about 10 times as much work as an i3!
  • An AMD GPU, of the 4000 series or later (so it's OpenCL capable), with the 10.10 or later Catalyst drivers, of the APP variety. A mid-range Radeon 5770 does 2-3 times as much work as an i3. Higher-end AMD GPUs are likely to require a fast CPU core as well for optimum performance. Running 64-bit Linux can help mitigate this problem.
  • "Use ATI GPU" and "Use NVIDIA GPU" set to "yes" in your PrimeGrid preferences.
  • Windows users should make sure they have msvcrt90.dll. You can get it by installing the .NET 3.5 runtime. (This might be fixed for the AMD apps; not sure.)
  • Users with multiple GPUs should add a cc_config.xml file.

What would help a little, but isn't absolutely necessary:
  • A 64-bit OS on a 64-bit processor will run CPU work 30% faster than a 32-bit OS. It matters very little for GPU work, except for some fast AMD GPUs in Linux or MacOS.
  • Overclocking a little. This project produces less heat than the LLR challenge from last month, and it's more tolerant of errors - except on Nvidia GPUs, so watch out! AMD GPUs will have their errors auto-corrected this year! But the auto-correction process requires CPU time, so there is a point of diminishing returns.
  • This section isn't entirely tested. The following app_config.xml may slightly help Nvidia cards, 400-series or later, if you have BOINC 7.2.39 or later:
    Code:
    <app_config>
      <app>
        <name>pps_sr2sieve</name>
        <max_concurrent>9</max_concurrent>
      </app>
      <app_version>
        <app_name>pps_sr2sieve</app_name>
        <plan_class>cuda23</plan_class>
        <cmdline>-m64</cmdline>
      </app_version>
    </app_config>
    The same thing can be done with app_info.xml, but it's more difficult to write such a file.

What won't help (but won't hurt either):
  • A large amount of RAM.
  • Any older video cards or Intel video cards - though you might want to use them for your display while newer cards crunch. The new OpenCL apps can actually run on Intel; convincing BOINC to use them is the hard part.

This challenge starts on December 18, at 23:03 UTC and runs until the Winter Solstice three days later. Only WUs downloaded after the start of the challenge and completed before the end of the challenge will be counted. Should you not be able to be in front of one or more computers at the start time, either set their queue level to zero (0) days or try one of the scheduled start methods from the strategy thread (for Linux or Windows). Network connection time options may also be useful - though they depend on your computer's clock being correct, so set the start time a few minutes late.

Welcome and good luck to all! :thumbsup:
 
Last edited:

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,284
3,905
75
Will folding during the PrimeGrid race count for the F@H race?
 

petrusbroder

Elite Member
Nov 28, 2004
13,343
1,138
126
Well, in some previous years we have decided on a case-by-case-basis; most of the time it did not make much of a difference. Last year we have counted the period into the race as the difference between the teams was minimal ...
Some years ago, all the folders (except one or two) switched over to the PrimeGrid race ..., has not happended since, though.
I'll keep all posted.
 

Plimogz

Senior member
Oct 3, 2009
678
0
71
The rough performance estimates in the OP has me wondering:

A GTX460 is 10x an i3 and a 5770 is 2x-3x -- but in most applications (games) the performance disparity between those cards isn't that big.

Is the nVidia implementation much more efficient than the AMD one?
 

GLeeM

Elite Member
Apr 2, 2004
7,199
128
106
<app_config>
<app_version>
<app_name>pps_sr2sieve</app_name>
<plan_class>cuda23</plan_class>
<cmdline>-m64</cmdline>
</app_version>
</app_config>
This didn't work for me, said something like wrong app_name?
What does "-m64" do?

On GTX 560Ti core 448, two WUs did not go faster, you were right.
It will get ~330,000 ppd running one WU.
 

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,284
3,905
75
This didn't work for me, said something like wrong app_name?
I have no idea why that wouldn't work. I have "<app_name>pps_sr2sieve</app_name>" in my client_state.xml, followed closely by "<plan_class>cuda23</plan_class>".

What does "-m64" do?
Adjusts the number of threads that run on the GPU at once.

Edit: Maybe it needs an <app> section before the <app_version> section?
Code:
  <app>
    <name>pps_sr2sieve</name>
    <max_concurrent>9</max_concurrent>
  </app>
 
Last edited:

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,284
3,905
75
Is the nVidia implementation much more efficient than the AMD one?
Yes, to varying degrees. Fermi and later do true 32-bit multiplies; plus there's a CUDA instruction that really helps. But AMD GCN is now significantly improved from last year.
 

GLeeM

Elite Member
Apr 2, 2004
7,199
128
106
Maybe it needs an <app> section before the <app_version> section?
There were no errors with the following:

<app_config>
<app>
<name>pps_sr2sieve</name>
<plan_class>cuda23</plan_class>
<cmdline>-m64</cmdline>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.10</cpu_usage>
</gpu_versions>
</app>
</app_config>

How can I tell if it is using 64 threads?
 

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,284
3,905
75
How can I tell if it is using 64 threads?
That's a bug in my software: You can't. :twisted:

I don't think that's going to do it, though. Are you on BOINC 7.2.39 or later?
 

GLeeM

Elite Member
Apr 2, 2004
7,199
128
106
Are you on BOINC 7.2.39 or later?
Yes, 7.2.42

How much of a time difference per WU should I see if "<cmdline>-m64</cmdline>" is working?

Running two WUs on the GTX 780 only made a few seconds difference.
 

biodoc

Diamond Member
Dec 29, 2005
6,271
2,238
136
i'll pass on this race. We're too far behind on the FAH holiday race.
 

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,284
3,905
75
Running two WUs on the GTX 780 only made a few seconds difference.
Roughly 7%, I think. Looks like running two at once isn't effective.

Edit: Does the updated xml file in the OP work for you now?
 
Last edited:

GLeeM

Elite Member
Apr 2, 2004
7,199
128
106
Edit: Does the updated xml file in the OP work for you now?
No, I get the following:

12/15/2014 7:52:44 PM | PrimeGrid | Entry in app_config.xml for app 'pps_sr2sieve', plan class 'cuda23' doesn't match any app versions

?

EDIT: now after changing back to the file that showed no errors ... it shows the same error
 
Last edited:

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,284
3,905
75
And we're off! I'm keeping most of my CPUs in reserve on F@H. But my new GPU is on this.
 

geecee

Platinum Member
Jan 14, 2003
2,383
43
91
I'm actually doing the reverse. I'm giving you most of my CPUs, and using my primary GPU on folding. I do have an old nvidia GPU that wasn't contributing to the folding race at all, and I've put that on PG as well.
 

Plimogz

Senior member
Oct 3, 2009
678
0
71
I'm trying to get openCL GPU work, but the server is apparently in no hurry to send me any.
 

Sgraffite

Member
Jul 4, 2001
88
0
66
I turned on a GPU for this, is there a stats page for this contest somewhere?

Nevermind, I found the team and user stats.
 
Last edited:

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,284
3,905
75
Day 1 stats:

Rank___Credits____Username
30_____2592299____Lane42
47_____1712468____GLeeM
117____610151_____zzuupp
132____562957_____Joe
152____482053_____Sgraffite
238____276422_____wayliff
321____151695_____SlangNRox
322____148324_____Ken_g6
698____6742_______geecee

Rank__Credits____Team
6_____18742760___Team 2ch
7_____7146520____BOINC@MIXI
8_____6984712____BOINC.Italy
9_____6543111____TeAm AnandTech
10____6398158____Canada
11____6206011____[H]ard|OCP
12____5828459____Team China

9th is very good - especially with another race going!

Anyone ever get AMD/Open CL tasks?
MSI R9 290 4GB
I've been getting "atiPPSsieve" tasks on my 7950. Are these Open CL?
And that's why you're not getting OpenCL tasks, in a roundabout way. Basically, the old BOINC server on PrimeGrid doesn't know about OpenCL. Even though it's serving OpenCL work, it thinks it's not OpenCL. Unfortunately, the BOINC client does know about OpenCL, thinks it's not getting it, and thus refuses to serve it to R9 290s. You can get around this with an app_info.xml file. I'm just not sure how to configure it.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |