1) Message boards : Number crunching : Strange behavior running (2) projects (Message 4599)
Posted 17 days ago by Profile nexiagsi16v
Post:
In my system are 3x GPUs and a Ryzen 5. I took the temperature from the CPU to regulate the cooling system...it is the last in line of the heatproducers and the only way to conect the cooling with a temperature.
If only the GPUs are crunching, the temp is high at the GPUs, but the CPU is under 45°C. So the cooling system need finetuning, e.g. 45°C on the CPU -> 60% coolingpower for 65°C on the GPUs. The right setting you only get by try it step by step.
If only the CPU is crunching and you want 60°C on it, 60% cooling power is too much. 50% or less is enough. With the old setting, you never get under 60% cooling power for 60°C.

If you crunching always universe with the same workload, the BIOS-setting works fine. But if you switch the workload/projekts or during the Pentathlon or other challenges, you have to change the cooling settings very often and so you have always restart the system. Under windows, the finetuning is done in seconds.
2) Message boards : Number crunching : Strange behavior running (2) projects (Message 4597)
Posted 18 days ago by Profile nexiagsi16v
Post:
I think that is normal. Linux is the better OS for Universe. Look at WUprop...linux 64bit vs. win10 64bit...linux is 3 or 4x faster.
http://wuprop.boinc-af.org/results/projet.py?projet=Universe%40Home&application=Universe+BHspin+v2 for Universe BHspin v2.

A way for you is maybe, Win 10 as the OS and linux in a VM.

The drivers for GPUs under windows are better/newer/updated. Another thing is, i have a costum watercoolsystem. Under Linux i can´t regulate the pump and the fanspeeds. All on 100% and that is not what i want. ;-) So Linux in a VM and with the main Win10 OS, i can control everything for the GPUs and coolingsystem.
3) Message boards : News : No tasks (Message 4557)
Posted 29 Nov 2020 by Profile nexiagsi16v
Post:
Best wishes and take the time for a well recovery.
4) Message boards : News : New ULX's (Message 4077)
Posted 19 Mar 2020 by Profile nexiagsi16v
Post:
Better you stop ULX 0.12 and send tommorow only a few number out for testing.

<core_client_version>7.14.2</core_client_version>
<![CDATA[
<stderr_txt>
20:46:12 (4040): called boinc_finish(0)

</stderr_txt>
<message>
upload failure: <file_xfer_error>
<file_name>universe_ulx_502_12608_20000_1-999999_650000_2_r524705463_1</file_name>
<error_code>-131 (file size too big)</error_code>
</file_xfer_error>
</message>
]]>

WU was downloaded 19 Mar 2020, 17:24:15 UTC for Win 10 and Ryzen 3600.
5) Message boards : Cafe : BOINC Pentathlon 2018 (Message 2858)
Posted 4 May 2018 by Profile nexiagsi16v
Post:
....and Universe is a part of it, again!

Happy to see this. I hope Krzysztof is ready!


I think yes...he put WU after WU on the Server. I hope he has a fire extinguisher close to the server for tomorrow. :))
6) Message boards : Cafe : BOINC Pentathlon 2018 (Message 2848)
Posted 2 May 2018 by Profile nexiagsi16v
Post:
....and Universe is a part of it, again!
7) Message boards : News : BHDB application (Message 2693)
Posted 11 Mar 2018 by Profile nexiagsi16v
Post:
Great Job. This Projekt is on a high level und one of the best! The 0.03 are very short (an old j2900 with Linux need 11 - 13 minutes) and until now, no one is invalid.
8) Message boards : News : BHDB application (Message 2661)
Posted 10 Mar 2018 by Profile nexiagsi16v
Post:
Some of the WUs running fine, with the same specs of maximum discusage.
The WUs that crashing on my system, crash on other systems, too.

I have now a WU with 100% and one data-file were only 135MB. I can´t look so fast for the other file, because the WU at this time was at 100% and next moment uploaded ... :-/

https://universeathome.pl/universe/result.php?resultid=33277772

The question is, why some WUs need so much diskspace? Can you resend some of this invalid with higher limit of allowed discuse? If they become valid, we find a way...if they become invalid (not by error discusage), theres a problem with the WUs and they never become valid.

Tschau Norman
9) Message boards : News : BHDB application (Message 2647)
Posted 10 Mar 2018 by Profile nexiagsi16v
Post:
I hope...Linux the same Problem now.

2 good...6 bad

https://universeathome.pl/universe/result.php?resultid=33258694

<core_client_version>7.4.23</core_client_version>
<![CDATA[
<message>
Maximum disk usage exceeded
</message>
<stderr_txt>

</stderr_txt>
]]>

Tschau Norman
10) Message boards : News : BHDB application (Message 2645)
Posted 10 Mar 2018 by Profile nexiagsi16v
Post:
Okay, i put a fx8370 and 6300 for the night on Universe....no fast machines, but more than no machines. :)

We will see in the morning the results...good night from Germany!
11) Message boards : News : BHDB application (Message 2643)
Posted 10 Mar 2018 by Profile nexiagsi16v
Post:
On my Win7 the WUs break up around 60-70%.

https://universeathome.pl/universe/results.php?hostid=921

<core_client_version>7.6.33</core_client_version>
<![CDATA[
<message>
Disk usage limit exceeded
</message>
<stderr_txt>

</stderr_txt>
]]>


My Diskspacelimit for Boinc is higher than the WUs needed....so it can´t be the problem. When i look into the Boinc-dir, /Slots....the files "Data1.dat2" and "Data1.dat3" go up to 3xx MB. Both files have the same size. Under Linux this files under 200MB and running fine to 100%.
12) Message boards : Number crunching : extreme long wu's (Message 2367)
Posted 20 Sep 2017 by Profile nexiagsi16v
Post:
250h for nothing....this WU stands at 0,1%.
WU http://universeathome.pl/universe/result.php?resultid=26112133
work packet http://universeathome.pl/universe/workunit.php?wuid=11376308







Copyright © 2021 Copernicus Astronomical Centre of the Polish Academy of Sciences
Project server and website managed by Krzysztof 'krzyszp' Piszczek