PrimeGrid
Please visit donation page to help the project cover running costs for this month

Toggle Menu

Join PrimeGrid

Returning Participants

Community

Leader Boards

Results

Other

drummers-lowrise
1) Message boards : Number crunching : New Badges (Updated 2017) (Message 106539)
Posted 1025 days ago by Robert Hoffman
Happy April Fool's Day to you and yours as well!

Phooey, beaten by 1998golfer.
2) Message boards : Number crunching : Badges II (Message 105175)
Posted 1065 days ago by Robert Hoffman
Courtesy of TdP:
3) Message boards : Generalized Cullen/Woodall prime search : Benchmarks for GCW Sieve (Message 100366)
Posted 1180 days ago by Robert Hoffman
Hyper Threading. It increases the individual time it takes for each task to complete, though it makes up for it by doing more tasks at once.


Didn't realize it was that big of a hit in performance for sieving tasks. I know that (compared to itself) it will run faster if I turn HT off and run 3/4 tasks.

But given that I'm not doing a full 8/8 tasks, a 23% faster clock (the i5 is actually running mid turbo @ 3.4) and slightly faster RAM (1866 vs 16oo), I really expected the i7 to be much faster on the individual tasks not just the overall throughput.

The main gripe of course is that this is work computer (off the shelf Lenovo) vs home computer (custom build, lovingly hand assembled from specifically chosen component parts) and the home computer should be absolutely and unequivocally crushing the work computer.


I could swear I remember reading at some point that the amount of cache played a factor with sieving...

My ancient C2Q has 12Mb of total cache and is now taking 7062 seconds per task on average with all 4 cores active. My I7-3770s have 8Mb. I'm running flat out with those, 8 threads each, and the best I can get is 8996 and 9084 seconds (average of last 20 tasks), Win7 vs. Win10 respectively. The Windows 10 has slightly faster memory as well.

I'll disable HT on the Windows 10 system once the current batch finishes and see what happens.
4) Message boards : Generalized Cullen/Woodall prime search : Benchmarks for GCW Sieve (Message 100161)
Posted 1186 days ago by Robert Hoffman
AMD Phenom II X6 1090T (3.2GHz), Windows 7 Pro , x64, DDR3-1600 - 3 simultaneous. Average is 10270 seconds.

Core 2 Quad Q9550 (2.83GHz), Ubuntu 16.04 x64, DDR2-800 - 4 simultaneous. Average is 7600 seconds.

I7-3770 (3.4GHz), Windows 7 Pro, x64, DDR3-1333 - 8 simultaneous. Average is 9911 seconds.

I7-3770 (3.4GHz), Windows 10 Pro, x64, DDR3-1600 - 8 simultaneous. Average is 9849 seconds.



5) Message boards : Number crunching : Boinc goes off at night (Message 97597)
Posted 1259 days ago by Robert Hoffman
Monthly Windows Updates came out yesterday, so depending on your configuration the computer may have installed them and rebooted automatically.
6) Message boards : Project Staging Area : 2015 PRPNet March Challenge (Message 83567)
Posted 1789 days ago by Robert Hoffman
I can only assume (..!) that the tasks are correct in that I earn credit both on PRPNet as well as on Primegrid via PSA credits.


Authoritative answer: That assumption is incorrect.



So noted, thank you for clarifying.

-Rob
7) Message boards : Project Staging Area : 2015 PRPNet March Challenge (Message 83561)
Posted 1790 days ago by Robert Hoffman
Are we sure that llrCUDA produces correct results? It's never been "live" on BOINC. Question spans various card/driver/OS combinations.

PRPnet work isn't double-checked (except for when a prime is "found", it would be checked at the Top 5000 site).

--Gary


Gary,

I can only assume (..!) that the tasks are correct in that I earn credit both on PRPNet as well as on Primegrid via PSA credits.

http://u-g-f.de/PRPNet/user_search?uid=Robert_Hoffman



Unless I am missing something, llrCUDA does not appear to obey the gpuaffinity settings in the prpclient. If I try to run on both GPUs in a dual-GPU system, I only end up running two instance on the device 0 GPU. Anyone have any luck with getting it to run on both GPUs?


Also, I think it is optimal to leave one core free for llrCUDA. You shouldn't need more than that per instance of llrCUDA running. This also works well in hyperthreading, as I can run 4 cores at 95% or more of full speed on another application while allowing llrCUDA to interact with the 4 free "cores" from the hyperthread and not take any performance hit itself (obviously, in Windows at least, the CPU applications are sometimes switching to the hyperthread since core assignment is variable in this OS setting without substantial micromanagement).




Scott,

I am finding that the +1 tasks only use a fraction of a CPU, while the -1 use a full CPU in addition to the GPU. I do have two GPUs in this system, but only the GTX 560ti is actively used for tasks. I have a Radeon HD5670 that is the primary display, although it has been used for some PPS-Sieve work in the past. I have not manually specified a GPU in the prpclient.ini file.

-Rob
8) Message boards : Project Staging Area : 2015 PRPNet March Challenge (Message 83531)
Posted 1791 days ago by Robert Hoffman
I finished 2 GPU tasks this afternoon and received credit for them...

They ran simultaneously, with an unintended reboot in the middle of the run, and took an average of 18230 seconds. When run singly, they average 8897 seconds on my stock GTX 560ti.

-Rob
9) Message boards : Project Staging Area : PRPNet 5.3.1 (Message 79390)
Posted 1960 days ago by Robert Hoffman
I'm noticing an issue with my client. I'm currently running 100% Wieferich with the following setting:

server=WIEFERICH:100:25:prpnet.primegrid.com:13000


To the best of my knowledge this should mean I get 25 work units, process 25 workunits and then return 25 workunits. However, what happens is the server (thinks it) sends 25, I get 24, I process 24 and I return 24. The 25th just disappears into thin air. You can check pending work and see that I have WU's pending with an interval of about 30 minutes.

I'm running a fully patched Win7 installation.


Yes, it's something that's been around a bit. I run 20 with no issues.

See here: http://www.primegrid.com/forum_thread.php?id=5157&nowrap=true#67531
10) Message boards : Proth Prime Search : ATI sieve app turned off; Help needed. (Message 77971)
Posted 2022 days ago by Robert Hoffman
./tpsieve-cl-boinc-x86-mac -p13120716e9 -P13120725e9 -k5 -K9999 -n6000000 -N9000000 -T -M2 -c 60
tpsieve version cl-0.2.5a beta (testing)
Compiled Jul 6 2014 with GCC 4.2.1 (Based on Apple Inc. build 5658) (LLVM build 2336.9.00)
nstart=6000000, nstep=41
tpsieve initialized: 5 <= k <= 9999, 6000000 <= n < 9000000
nstep changed to 32
CL setup complete.
cthread_count = 16384
13120716307705079 | 2959*2^8094354+1
13120716380747377 | 8859*2^7847166+1
13120716412325987 | 7211*2^7702451+1
13120716843190109 | 2167*2^7951358+1
13120716846657643 | 7003*2^8206351-1
13120717075819579 | 5089*2^8180582+1
13120717160591431 | 7745*2^8765661+1
13120717284827623 | 8907*2^8826378+1
13120717824685699 | 4613*2^7552946-1
13120717861537051 | 5537*2^7081032-1
13120718815920467 | 2501*2^8550910-1
13120718930066591 | 3699*2^7482137-1
13120719057141307 | 6089*2^8973032-1
13120719059517079 | 2445*2^7778847-1
13120719477216161 | 6435*2^8752766+1
13120719544063997 | 2985*2^7826085+1
13120719661679807 | 8253*2^8382530-1
13120719782159167 | 1543*2^7389775-1
13120719966492419 | 3749*2^6399636-1
13120720493876977 | 6645*2^7538049+1
13120720553340431 | 2055*2^8038331+1
13120720607106341 | 8311*2^7968572+1
13120720635031169 | 5679*2^8994161+1
13120720662833737 | 5469*2^6562478+1
13120720682005019 | 1935*2^7473932-1
13120721017491301 | 3531*2^6213937-1
13120721212413691 | 8341*2^6170513-1
13120721663769431 | 3585*2^6907848+1
13120721691198937 | 6395*2^7988954-1
13120721797389533 | 8075*2^8648682-1
13120721851239341 | 8595*2^6216665-1
13120721947519033 | 3327*2^6867165-1
13120722319646957 | 5003*2^6419493+1
13120722440108203 | 5707*2^6807554+1
13120722557491823 | 1731*2^8986587+1
13120722686753867 | 8135*2^7034866-1
13120723071996931 | 4241*2^7534559+1
13120723189240451 | 1119*2^6735109-1
13120723194075703 | 2055*2^6453532+1
13120723256598527 | 8165*2^8191313+1
13120723258653649 | 4355*2^7639969+1
13120723434822539 | 8535*2^6601016+1
13120723717933309 | 9707*2^7180267+1
13120723954016033 | 4685*2^8440045+1
13120723983397483 | 5737*2^6363638+1
13120724045408101 | 2565*2^6057669+1
13120724075161099 | 6163*2^7807643-1
13120724084494907 | 3081*2^6075501+1
13120724364906871 | 4281*2^8009816+1
13120724414854831 | 7821*2^6744179+1
13120724617851359 | 9975*2^7303461-1
13120724673389471 | 2361*2^6415555+1
13120724738634179 | 1485*2^6690363+1
Found 53 factors

stderr.txt
07:46:30 (1233): Can't open init data file - running in standalone mode
Sieve started: 13120716000000000 <= p < 13120725000000000
Thread 0 starting
Detected 64 multiprocessors (320 SPUs) on device 0.
Device 0 is a NVIDIA GeForce 8600M GT.
GCN device detected; use -m1 --vecsize=4 to undo effect

Thread 0 completed
Sieve complete: 13120716000000000 <= p < 13120725000000000
count=242515220,sum=0x2dca915341c02724
Elapsed time: 57581.96 sec. (0.34 init + 57581.62 sieve) at 156303 p/sec.
Processor time: 5335.63 sec. (0.35 init + 5335.28 sieve) at 1686920 p/sec.
Average processor utilization: 1.02 (init), 0.09 (sieve)
23:46:12 (1233): called boinc_finish


Next 10 posts
[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2020 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 1.99, 2.50, 2.62
Generated 21 Jan 2020 | 23:03:39 UTC