mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Hardware > GPU Computing

Reply
 
Thread Tools
Old 2014-07-16, 17:26   #1
sixblueboxes
 
sixblueboxes's Avatar
 
Feb 2013
Charlotte, North Carolina

31 Posts
Default NVIDIA Quadro K4000 speed results benchmark

I've read the FAQ and the PDF guide and I have some questions.

I have an NVIDIA Quadro K4000 and I can't seem to find much benchmark information on it. Here's some output from two exponents I've run LL tests on so far:

Code:
Starting M86243 fft length = 4608
Iteration 5000 M( 86243 )C, 0xd441a52cc693a4eb, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4157 ms/iter, ETA 0:33)
Iteration 10000 M( 86243 )C, 0x23992ccd735a03d9, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4025 ms/iter, ETA 0:30)
Iteration 15000 M( 86243 )C, 0xff29912e0d68a645, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.3989 ms/iter, ETA 0:27)
Iteration 20000 M( 86243 )C, 0x89c58d63ebee7ad1, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4012 ms/iter, ETA 0:26)
Iteration 25000 M( 86243 )C, 0x52a71c358eeb3457, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4014 ms/iter, ETA 0:24)
Iteration 30000 M( 86243 )C, 0x8ad9ad7af5b51d09, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.3991 ms/iter, ETA 0:21)
Iteration 35000 M( 86243 )C, 0x5b0123e427f29a72, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4015 ms/iter, ETA 0:20)
Iteration 40000 M( 86243 )C, 0xeed70124ff3b4f5a, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.3989 ms/iter, ETA 0:17)
Iteration 45000 M( 86243 )C, 0xe4ee04df8f36044b, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4000 ms/iter, ETA 0:16)
Iteration 50000 M( 86243 )C, 0x6ef44d2b23c538e1, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4017 ms/iter, ETA 0:14)
Iteration 55000 M( 86243 )C, 0x30721775a8cc2eab, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4001 ms/iter, ETA 0:12)
Iteration 60000 M( 86243 )C, 0x76f20516c9858691, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4022 ms/iter, ETA 0:10)
Iteration 65000 M( 86243 )C, 0x0ee5b3cbbe65ae13, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4003 ms/iter, ETA 0:08)
Iteration 70000 M( 86243 )C, 0x1c98576ef37a22df, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4113 ms/iter, ETA 0:06)
Iteration 75000 M( 86243 )C, 0x875f23fbc68f3847, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4424 ms/iter, ETA 0:04)
Iteration 80000 M( 86243 )C, 0x6809f7b9c9f1e33d, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4005 ms/iter, ETA 0:02)
Iteration 85000 M( 86243 )C, 0xde8f08c8df8f7495, n = 4608, CUDALucas v2.03 err = 0.0122 (0:02 real, 0.4036 ms/iter, ETA 0:00)
M( 86243 )P, n = 4608, CUDALucas v2.03



Iteration 4215000 M( ........ )C, 0x17dd5e869cf760d3, n = 3932160, CUDALucas v2.03 err = 0.0826 (2:02 real, 24.3668 ms/iter, ETA 427:46:24)
Iteration 4220000 M( ........ )C, 0x2d714f472bc32c9a, n = 3932160, CUDALucas v2.03 err = 0.0826 (2:02 real, 24.3593 ms/iter, ETA 427:36:23)
Iteration 4225000 M( ........ )C, 0x484e0b71c0395242, n = 3932160, CUDALucas v2.03 err = 0.0826 (2:01 real, 24.3714 ms/iter, ETA 427:47:10)
Iteration 4230000 M( ........ )C, 0x7455eeed6495d491, n = 3932160, CUDALucas v2.03 err = 0.0826 (2:02 real, 24.4089 ms/iter, ETA 428:24:35)
Iteration 4235000 M( ........ )C, 0x535bf19776853a8e, n = 3932160, CUDALucas v2.03 err = 0.0826 (2:01 real, 24.3561 ms/iter, ETA 427:26:59)
Iteration 4240000 M( ........ )C, 0x3e83b35304e5da1a, n = 3932160, CUDALucas v2.03 err = 0.0826 (2:02 real, 24.5831 ms/iter, ETA 431:23:54)
Iteration 4245000 M( ........ )C, 0x7d4fd2f79217477a, n = 3932160, CUDALucas v2.03 err = 0.0826 (2:02 real, 24.5553 ms/iter, ETA 430:52:36)
Iteration 4250000 M( ........ )C, 0xae3ddf2e36775c94, n = 3932160, CUDALucas v2.03 err = 0.0826 (2:03 real, 24.5392 ms/iter, ETA 430:33:38)
Iteration 4255000 M( ........ )C, 0xc186c15482785129, n = 3932160, CUDALucas v2.03 err = 0.0826 (2:03 real, 24.5423 ms/iter, ETA 430:34:54)
Iteration 4260000 M( ........ )C, 0xa9a712f4cd548353, n = 3932160, CUDALucas v2.03 err = 0.0878 (2:06 real, 25.1284 ms/iter, ETA 440:49:44)
Iteration 4265000 M( ........ )C, 0x4c1b2f4cf3a408df, n = 3932160, CUDALucas v2.03 err = 0.0878 (2:02 real, 24.3625 ms/iter, ETA 427:21:31)
Iteration 4270000 M( ........ )C, 0xe1c84663109f99a8, n = 3932160, CUDALucas v2.03 err = 0.0878 (2:02 real, 24.3674 ms/iter, ETA 427:24:39)
Iteration 4275000 M( ........ )C, 0xec8c21a537c82f5a, n = 3932160, CUDALucas v2.03 err = 0.0878 (2:02 real, 24.4065 ms/iter, ETA 428:03:43)
Iteration 4280000 M( ........ )C, 0xc3d6022cbfe47db9, n = 3932160, CUDALucas v2.03 err = 0.0878 (2:02 real, 24.5533 ms/iter, ETA 430:36:10)
Iteration 4285000 M( ........ )C, 0xae1ff4acc09453db, n = 3932160, CUDALucas v2.03 err = 0.0878 (2:03 real, 24.6085 ms/iter, ETA 431:32:14)
Iteration 4290000 M( ........ )C, 0x4dae1a2a5ecedd68, n = 3932160, CUDALucas v2.03 err = 0.0878 (2:02 real, 24.3771 ms/iter, ETA 427:26:41)
Iteration 4295000 M( ........ )C, 0x39ad5a2fa37d46cc, n = 3932160, CUDALucas v2.03 err = 0.0878 (2:02 real, 24.3737 ms/iter, ETA 427:21:08)
Iteration 4300000 M( ........ )C, 0x7fc062edc96f3d34, n = 3932160, CUDALucas v2.03 err = 0.0878 (2:03 real, 24.6717 ms/iter, ETA 432:32:33)
Iteration 4305000 M( ........ )C, 0x6e130f7db734c5a1, n = 3932160, CUDALucas v2.03 err = 0.0878 (2:02 real, 24.4130 ms/iter, ETA 427:58:25)
The first part is the complete test on the smallest number it does, and the second part is an excerpt of the output from working on one of my manual assignments. I've noticed that the ms/iter are much higher for the higher number, which I suppose makes sense. Can I get it to go faster? Is anyone else running this card that can comment or post their output? Any insight will be greatly appreciated.

Last fiddled with by sixblueboxes on 2014-07-16 at 17:34
sixblueboxes is offline   Reply With Quote
Old 2014-07-16, 21:06   #2
TheMawn
 
TheMawn's Avatar
 
May 2013
East. Always East.

6BF16 Posts
Default

As far as I know, the professional series of GPU's aren't actually very fast. They're reliable as hell, though.

What is the second exponent? Even just a ballpark like how many million?
TheMawn is offline   Reply With Quote
Old 2014-07-16, 22:08   #3
Batalov
 
Batalov's Avatar
 
"Serge"
Mar 2008
Phi(4,2^7658614+1)/2

41×229 Posts
Default

Quote:
Originally Posted by TheMawn View Post
What is the second exponent? Even just a ballpark like how many million?
Can't you easily determine it from the data presented? Think for a few seconds...

I'll give you a hint: ETA_in_seconds*1000/millisec_per_iteration + num_iter_already_done = fairly precise number (precision = +- 1000)
Batalov is offline   Reply With Quote
Old 2014-07-17, 00:25   #4
firejuggler
 
firejuggler's Avatar
 
Apr 2010
Over the rainbow

43×59 Posts
Default

around 64 109 081 ? (24.5 ms by itération)
firejuggler is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
A dream, will stay a dream ( new Nvidia Quadro) firejuggler GPU Computing 0 2018-03-28 16:02
Nvidia 364.xx drivers returning incorrect results UBR47K GPU Computing 2 2016-06-11 22:38
Strange benchmark results AlTonno15 Information & Answers 3 2013-01-29 02:23
Wrong CPU speed reported in mprime benchmark James Heinrich Software 5 2009-06-13 12:56
no speed gain in Prime95's benchmark when overclocking forcemaker Hardware 8 2006-04-02 15:18

All times are UTC. The time now is 01:13.

Tue Apr 13 01:13:23 UTC 2021 up 4 days, 19:54, 1 user, load averages: 1.90, 2.50, 2.53

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.