![]() |
![]() |
#12 | |
(loop (#_fork))
Feb 2006
Cambridge, England
13·491 Posts |
![]() Quote:
having some of its memory bandwidth stolen by worker2. This is consistent with your measurements of worker2 being much slower than worker1 and both workers being slower than before. |
|
![]() |
![]() |
![]() |
#13 |
Aug 2020
2510 Posts |
![]()
I will once I get home.
I rebooted the PC this morning and noticed the time/interactions have changed. Both workes went to 7, 8 ish and few mins later went to 12 again. These two 2690 are liquid cooled. Temp does not go beyond 70C on each cpu. Last fiddled with by wagner85 on 2020-11-17 at 14:25 |
![]() |
![]() |
![]() |
#14 | |
Aug 2020
52 Posts |
![]() Quote:
I have just replaced the old cpus with the new ones. How do I confirm that hypothesis ? I will paste here the prime.txt for both servers once I get home. Would I solve this by starting a new p95 folder and move the files related to my current work to new folder? |
|
![]() |
![]() |
![]() |
#15 |
"/X\(‘-‘)/X\"
Jan 2013
3×977 Posts |
![]()
Remove any Affinity= lines from local.txt while Prime95/mprime isn't running.
If that doesn't help, I'd verify that each memory channel is populated evenly. I don't know if you have 8 or 16 DIMM motherboards. Last fiddled with by Mark Rose on 2020-11-18 at 00:37 |
![]() |
![]() |
![]() |
#16 |
Aug 2020
52 Posts |
![]()
This is the prime.txt for 2620s
V24OptionsConverted=1 V30OptionsConverted=1 WGUID_version=2 StressTester=0 UsePrimenet=1 DialUp=0 V5UserID=wagner85 PRPGerbiczCompareIntervalAdj=1 Priority=1 DaysOfWork=3 RunOnBattery=1 WorkPreference=151 [PrimeNet] Debug=2 ProxyHost= UploadRateLimit=10000 UploadStartTime=23 UploadEndTime=7:00 DownloadDailyLimit=10000 [Worker #1] [Worker #2] THIS IS THE LOCAL.TXT FOR 2620s OldCpuSpeed=2000 NewCpuSpeedCount=0 NewCpuSpeed=0 RollingAverage=1054 RollingAverageIsFromV27=1 WorkerThreads=2 CoresPerTest=6 ComputerGUID=63e51cf8d701208e2becb8153e90a4ae ComputerID=Mireia WorkerDiskSpace=400 Memory=28672 during 7:30-23:30 else 28672 MaxEmergencyMemory=7987 CertDailyCPULimit=10 CertWork=1 Pid=3411 SrvrUID=384350646 SrvrComputerName=135955582 SrvrPO2=1 SrvrPO3=3 SrvrPO4=28672 SrvrPO5=28672 SrvrPO6=450 SrvrPO7=1410 SrvrPO8=1 SrvrPO9=2 SrvrP00=4 LastEndDatesSent=1605591311 CertDailyRemainingLastUpdate=1605615511 CertDailyMBRemaining=10000 CertDailyCPURemaining=10 RollingHash=2626590820 RollingStartTime=1605608306 RollingCompleteTime=2753156 SrvrPO1=151 [Worker #1] [Worker #2] Last fiddled with by wagner85 on 2020-11-18 at 12:36 |
![]() |
![]() |
![]() |
#17 | |
Aug 2020
52 Posts |
![]() Quote:
24OptionsConverted=1 V30OptionsConverted=1 WGUID_version=2 StressTester=0 UsePrimenet=1 DialUp=0 V5UserID=wagner85 PRPGerbiczCompareIntervalAdj=1 Priority=10 DaysOfWork=3 RunOnBattery=1 WorkPreference=153 MinBenchFFT=16000 MaxBenchFFT=16332 BenchErrorCheck=0 BenchAllComplex=0 OnlyBench5678=0 BenchCores=16 BenchHyperthreads=0 BenchWorkers=2,4,8,16 AllBench=0 BenchTime=5 OutputIterations=10000 ResultsFileIterations=99999 DiskWriteTime=30 NetworkRetryTime=2 NetworkRetryTime2=70 DaysBetweenCheckins=1 NumBackupFiles=3 SilentVictory=0 MinTortureFFT=22 MaxTortureFFT=106 TortureMem=0 TortureTime=6 TortureWeak=0 [PrimeNet] Debug=2 ProxyHost= UploadRateLimit=1000 UploadStartTime=22:00 UploadEndTime=7:00 DownloadDailyLimit=100000 [Worker #1] [Worker #2] [Worker #3] [Worker #4] [Worker #5] [Worker #6] [Worker #7] [Worker #8] [Worker #9] [Worker #10] [Worker #11] [Worker #12] This is the LOCAL.txt for 2690s.TXT OldCpuSpeed=2900 NewCpuSpeedCount=0 NewCpuSpeed=0 RollingAverage=1057 RollingAverageIsFromV27=1 WorkerThreads=2 CoresPerTest=8 ComputerGUID=b77842ece9aca7608468641e1a60475a ComputerID=Massaranduba CertWork=1 Pid=3318 SrvrP00=4 LastEndDatesSent=1605664127 CertDailyRemainingLastUpdate=1605689449 CertDailyMBRemaining=100000 CertDailyCPURemaining=10 RollingHash=1424349509 RollingStartTime=1605682244 RollingCompleteTime=1402013 WorkerDiskSpace=400 Memory=28876 during 7:30-23:30 else 28876 MaxEmergencyMemory=7987 CertDailyCPULimit=10 SrvrUID=384350646 SrvrComputerName=1368227933 SrvrPO1=153 SrvrPO2=10 SrvrPO3=3 SrvrPO4=28876 SrvrPO5=28876 SrvrPO6=450 SrvrPO7=1410 SrvrPO8=1 SrvrPO9=2 [Worker #1] [Worker #2] [Worker #3] [Worker #4] [Worker #5] [Worker #6] [Worker #7] [Worker #8] [Worker #9] [Worker #10] [Worker #11] [Worker #12] Last fiddled with by wagner85 on 2020-11-18 at 12:33 |
|
![]() |
![]() |
![]() |
#18 |
Aug 2020
318 Posts |
![]()
Hey guys thanks for helping me in the troubleshooting! I have been busy these days and couldn’t research my problem further.
So I suspect the issue is on the PSU or in the cpus. So far mostly PSU. The reason: I measured my wattage consumption during prime usage. ( check the pdf for data). When the PC is idddle is consumes 110 w. After stating to crunch the numbers the power consumption goes to 530W. Right There is where I have the best performance around 7 ms/itr. A minute later the power consumption goes down to around 500 and the ms/itr goes a little bit upper 7.... And that keeps going on until the consumption stabilizes in 370 W and me/itr close do 16.... I could correlate the wattage consumption to the performance. See data attached. In order to verify if I was restrained by the PSU. I ran a small test using just one CPU instead of both. The results I got were 16 ms/itr. I did not check the power consumption. So if I was being restrained by PSU using two CPUs, I thought I would not be so restrained using just one cpu. Therefore I was expecting to find lower ms/itr. That did not happen!!! My power supply is rated for 500 W. I have a Z9PR-D12 MB. Before the cpu upgrades the consumption was around 330 W running p95. (2 x e5-2620_ 95W ea). Since then I removed all the 6 1u fans and added two water cooling systems. I tried to plug two 500W PSU. One supplying power to the MB and the other for both CPUs. Glad I did not burn my machine. Later I read how bad that can be. I also ran tests on a new p95 copy and I got similar ms/itr... around 16. So that leads me to believe it is not a config issue. Also changing the parameters in the BIOS had no shown no difference in ms/itr. As of now my 2620 machine consumes less power and works faster than my 2690 machine. Any thoughts? Thanks again for your suggestions. |
![]() |
![]() |
![]() |
#19 |
"Curtis"
Feb 2005
Riverside, CA
3×1,579 Posts |
![]()
If you're drawing 530W from a 500W power supply, your hardware will not last long. If you're somewhat lucky, the power supply will fail. If you're less lucky, it'll fail in a way that blows up the motherboard, or a CPU, or maybe a smoky mess when it goes...
Get a new power supply. One big enough to power your machine at less than 80% load- so, 700W minimum. |
![]() |
![]() |
![]() |
#20 | |
Aug 2020
52 Posts |
![]() Quote:
I agree! But that is just in the peak! The PC works around 370W. No more than that! And it has been running like that for 2 weeks now I believe. |
|
![]() |
![]() |
![]() |
#21 | |
Jun 2003
10011001100102 Posts |
![]() Quote:
watch "grep 'cpu MHz' /proc/cpuinfo" will show if CPU is throttling. But I have no solution if that is the case (apart from reinstalling / replacing cooler) |
|
![]() |
![]() |
![]() |
#22 |
Jun 2003
2·33·7·13 Posts |
![]() |
![]() |
![]() |
![]() |
Thread Tools | |
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
Lower Lip | R.D. Silverman | Soap Box | 10 | 2013-07-30 14:50 |
lower bounds on incomplete factorizations | J.F. | Factoring | 3 | 2008-06-14 18:58 |
Lower bounds for odd multiperfect numbers. | jchein1 | Math | 7 | 2006-11-26 13:29 |
How do I lower Prime95's CPU Usage? | Unregistered | Software | 21 | 2003-12-13 15:26 |
Where do I find undoc.txt to lower the usage? | Unregistered | Software | 3 | 2003-12-12 21:47 |