CPU time difference

Message boards : Number crunching : CPU time difference
Message board moderation

To post messages, you must log in.

1 · 2 · 3 · 4 · Next

AuthorMessage
Matthew@SETI@home

Send message
Joined: 14 Feb 16
Posts: 40
Credit: 9,278,146
RAC: 6
United States
Message 1793411 - Posted: 4 Jun 2016, 16:33:15 UTC

Re: http://setiathome.berkeley.edu/workunit.php?wuid=2173961623

How can two computers rated at roughly the same speed produce the same result, one consuming 8X the CPU time as the other?
ID: 1793411 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1793413 - Posted: 4 Jun 2016, 16:37:50 UTC - in response to Message 1793411.  

Re: http://setiathome.berkeley.edu/workunit.php?wuid=2173961623

How can two computers rated at roughly the same speed produce the same result, one consuming 8X the CPU time as the other?

Hm... very interesting case.
What about cache (L2 cache) size of your i3 and wingman's i5?
ID: 1793413 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1793422 - Posted: 4 Jun 2016, 17:02:16 UTC - in response to Message 1793411.  

Re: http://setiathome.berkeley.edu/workunit.php?wuid=2173961623

How can two computers rated at roughly the same speed produce the same result, one consuming 8X the CPU time as the other?

All CPU tasks on your i3-6100 are taking a very long time. You may want to use an app like CPUz to confirm that your CPU is running at its rated clock speed. If it isn't perhaps there could be a thermal issue.
If everything looks OK I would suggest suspending GPU processing and see if the CPU times improve.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1793422 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1793423 - Posted: 4 Jun 2016, 17:04:54 UTC - in response to Message 1793413.  

Re: http://setiathome.berkeley.edu/workunit.php?wuid=2173961623

How can two computers rated at roughly the same speed produce the same result, one consuming 8X the CPU time as the other?

Hm... very interesting case.
What about cache (L2 cache) size of your i3 and wingman's i5?

i3-6100 3MB
i5-2500 6MB
http://ark.intel.com/compare/52209,90729
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1793423 · Report as offensive
Matthew@SETI@home

Send message
Joined: 14 Feb 16
Posts: 40
Credit: 9,278,146
RAC: 6
United States
Message 1793427 - Posted: 4 Jun 2016, 17:19:40 UTC - in response to Message 1793422.  

If everything looks OK I would suggest suspending GPU processing and see if the CPU times improve.


How do you suspend GPU processing on just one machine of three?
ID: 1793427 · Report as offensive
Al Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Avatar

Send message
Joined: 3 Apr 99
Posts: 1682
Credit: 477,343,364
RAC: 482
United States
Message 1793433 - Posted: 4 Jun 2016, 17:47:36 UTC - in response to Message 1793427.  

Right click on the BOINC icon in the task tray down by the clock, and choose Snooze GPU tasks. That should do it.

ID: 1793433 · Report as offensive
Matthew@SETI@home

Send message
Joined: 14 Feb 16
Posts: 40
Credit: 9,278,146
RAC: 6
United States
Message 1793439 - Posted: 4 Jun 2016, 18:18:37 UTC - in response to Message 1793423.  
Last modified: 4 Jun 2016, 18:22:22 UTC

Ah, ok. Different definition of "snooze" from what I'm used to, which means automatic resume after x time. I'll try that if it comes to it. But I suspect it wouldn't survive a restart, so it wouldn't do as a long-term solution.

I haven't installed CPU-Z yet, but I did install Core Temp, which shows core temps around 60-65C with brief spikes as high as 75C. Is that high? If so, what should I do to address it? Take it to the shop and have them make sure the fan is working as designed? There is little if any noise, but I can feel warm air exiting the rear (very slowly). The computer is a 2-month-old Acer desktop.

What about cache (L2 cache) size of your i3 and wingman's i5?

Per Core Temp: L1 128KB, L2 512KB, L3 3MB
ID: 1793439 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14649
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1793441 - Posted: 4 Jun 2016, 18:22:18 UTC - in response to Message 1793439.  

'Snooze' is for a fixed interval of 1 hour.
ID: 1793441 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1793442 - Posted: 4 Jun 2016, 18:23:07 UTC - in response to Message 1793433.  
Last modified: 4 Jun 2016, 18:24:34 UTC

Right click on the BOINC icon in the task tray down by the clock, and choose Snooze GPU tasks. That should do it.

That will only cause GPU processing to stop for 1 hr. Using the menu option to select suspend GPU processing will remain active until changed again by the user.

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1793442 · Report as offensive
Matthew@SETI@home

Send message
Joined: 14 Feb 16
Posts: 40
Credit: 9,278,146
RAC: 6
United States
Message 1793447 - Posted: 4 Jun 2016, 18:34:40 UTC - in response to Message 1793442.  

Using the menu option to select suspend GPU processing will remain active until changed again by the user.

Got it, that does survive a restart. Thought I knew the BOINC manager better than that!
ID: 1793447 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1793461 - Posted: 4 Jun 2016, 19:36:23 UTC

do you have similar difference with MultiBeam tasks too?
ID: 1793461 · Report as offensive
Matthew@SETI@home

Send message
Joined: 14 Feb 16
Posts: 40
Credit: 9,278,146
RAC: 6
United States
Message 1793470 - Posted: 4 Jun 2016, 20:07:25 UTC - in response to Message 1793461.  
Last modified: 4 Jun 2016, 20:17:03 UTC

I don't know, I don't have many in the current tasks list that look like apples-to-apples comparisons. See for yourself: http://setiathome.berkeley.edu/results.php?hostid=7973906&offset=0&show_names=0&state=4&appid=29

Here's one that shows a 3X difference instead of 8X: http://setiathome.berkeley.edu/workunit.php?wuid=2174577485
ID: 1793470 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1793485 - Posted: 4 Jun 2016, 20:53:34 UTC - in response to Message 1793470.  

I don't know, I don't have many in the current tasks list that look like apples-to-apples comparisons. See for yourself: http://setiathome.berkeley.edu/results.php?hostid=7973906&offset=0&show_names=0&state=4&appid=29

Here's one that shows a 3X difference instead of 8X: http://setiathome.berkeley.edu/workunit.php?wuid=2174577485


Check if CPU operates on stock (3,7GHz) freq indeed and no iGPU tasks running.
If big difference still remain then it's smaller cache effect perhaps.
For MB it should be smaller than for AP (just as you see right now).
ID: 1793485 · Report as offensive
Stephen "Heretic" Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 20 Sep 12
Posts: 5557
Credit: 192,787,363
RAC: 628
Australia
Message 1793594 - Posted: 5 Jun 2016, 3:24:33 UTC - in response to Message 1793411.  

Re: http://setiathome.berkeley.edu/workunit.php?wuid=2173961623

How can two computers rated at roughly the same speed produce the same result, one consuming 8X the CPU time as the other?


. . HI,

. . I have a core i5-6400 and disabled crunching on the iGPU (same as yours a HD530) some time ago because it was crippling the output of this machine.

. . Your i3 will probably be a little slower than the i5 you were comparing it to but if you go your home page and in seti@home preferences turn of iGPU processing you will be doing yourself a favour. You may want to update some drivers as well. There may also be some other issues, how hot is the climate you are in? My machine is only running at 53 deg C with 3 CPU.s crunching and running an Nvidia graphics card. But as others have advised you running Seti on iGPU's can be problematical, it slows down the CPU performace surpisingly badly. When I was using the GPU my CPU tasks were taking 3.5 hours to run, when I stopped it the runtimes dropped to about 2 hours. I then did a Lunatics install and the times are now just over an hour per WU running under AVX.

. . Running just 3 CPU's my output has been far better than when I was running 3CPUs and the iGPU. Of course it is getting better again since I added the Nvidia graphics card :)
ID: 1793594 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 34744
Credit: 261,360,520
RAC: 489
Australia
Message 1793607 - Posted: 5 Jun 2016, 5:19:02 UTC

Yep, you can't compare i3 CPU times to i5 times as the i3 is 4 virtual cores made out of 2 physical cores whereas the i5 is just 4 pure cores, no comparison at all.

Cheers.
ID: 1793607 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1793678 - Posted: 5 Jun 2016, 14:52:17 UTC - in response to Message 1793607.  

Yep, you can't compare i3 CPU times to i5 times as the i3 is 4 virtual cores made out of 2 physical cores whereas the i5 is just 4 pure cores, no comparison at all.

Cheers.

Their i3 host is clearly having some other issue. My old i3-390M can complete work several times faster than what they are presently seeing.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1793678 · Report as offensive
Matthew@SETI@home

Send message
Joined: 14 Feb 16
Posts: 40
Credit: 9,278,146
RAC: 6
United States
Message 1793701 - Posted: 5 Jun 2016, 16:54:15 UTC - in response to Message 1793678.  
Last modified: 5 Jun 2016, 17:02:42 UTC

Core Temp shows Frequency as: 3688.27MHz (99.68 x 37.0). With that info is there any need to install CPU-Z?

After about a day, Core Temp now shows Max Core Temp as 79C, up from 75C after 30 minutes. Immediate temps still in the 60-65 range.

Any other suggestions as to how to proceed? I can't live with a significantly underperforming computer, but this problem exceeds my geek level.
ID: 1793701 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1793711 - Posted: 5 Jun 2016, 17:40:05 UTC - in response to Message 1793701.  

I can't live with a significantly underperforming computer, but this problem exceeds my geek level.

Try to limit number of cores BOINC use to 2 instead of 4.
How this will affect runtime?
ID: 1793711 · Report as offensive
Matthew@SETI@home

Send message
Joined: 14 Feb 16
Posts: 40
Credit: 9,278,146
RAC: 6
United States
Message 1793713 - Posted: 5 Jun 2016, 17:54:13 UTC - in response to Message 1793711.  
Last modified: 5 Jun 2016, 18:02:37 UTC

Try to limit number of cores BOINC use to 2 instead of 4.

This is by specifying "Use at most 50% of the CPUs" in Computing preferences?

How this will affect runtime?

I wouldn't know how to measure that with any accuracy, without running the same task before and after the change, which I assume is not possible. Host RAC might be useful, although it has yet to level out for this computer.
ID: 1793713 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6325
Credit: 106,370,077
RAC: 121
Russia
Message 1793714 - Posted: 5 Jun 2016, 18:07:06 UTC - in response to Message 1793713.  

Just do same comparison for few tasks as at thread start.
If the reason is hyperthreaded cores change will be quite significant.
ID: 1793714 · Report as offensive
1 · 2 · 3 · 4 · Next

Message boards : Number crunching : CPU time difference


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.