Posts by Senilix
|
1)
Message boards :
News :
MilkyWay@home Version 1.38 Released
(Message 65174)
Posted 19 Sep 2016 by  Senilix
Post: Something's not right with the opencl_ati_101 app. On my AMD Radeon R9 version 1.36 used to run 11 seconds, now version 1.38 needs more than 10 minutes (see http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=1791907717)!? Most of the time it's not even utilizing the graphic card but running solely on a full cpu core. Greetings, Senilix Edit: I compared the stdout files of a 1.36 job and a 1.38 job. Looks like 1.38 doesn't detect my gpu anymore... |
|
2)
Message boards :
Number crunching :
Validator is down
(Message 65030)
Posted 16 Aug 2016 by Senilix
Post: Tons of pending work units are piling up. |
|
3)
Message boards :
News :
New Separation Runs Started
(Message 57693)
Posted 28 Mar 2013 by Senilix
Post: Yep, there definitely is something wrong with these tasks. Mine was crashing after a split second. Here's the stderr output: <core_client_version>7.0.28</core_client_version>
<![CDATA[
<message>
Unzul�ssige Funktion. (0x1) - exit code 1 (0x1)
</message>
<stderr_txt>
<search_application> milkyway_separation 1.00 Windows x86 double </search_application>
Unrecognized XML in project preferences: max_gfx_cpu_pct
Skipping: 0
Skipping: /max_gfx_cpu_pct
Unrecognized XML in project preferences: apps_selected
Skipping: app_id
Skipping: /apps_selected
Unrecognized XML in project preferences: nbody_graphics_poll_period
Skipping: 30
Skipping: /nbody_graphics_poll_period
Unrecognized XML in project preferences: nbody_graphics_float_speed
Skipping: 5
Skipping: /nbody_graphics_float_speed
Unrecognized XML in project preferences: nbody_graphics_textured_point_size
Skipping: 250
Skipping: /nbody_graphics_textured_point_size
Unrecognized XML in project preferences: nbody_graphics_point_point_size
Skipping: 40
Skipping: /nbody_graphics_point_point_size
Error loading Lua script 'astronomy_parameters.txt': [string "number_parameters: 4..."]:1: '<name>' expected near '4'
Error reading astronomy parameters from file 'astronomy_parameters.txt'
Trying old parameters file
Integral area { 1819239215, 1769234796, 1008738314 } will overflow progress calculation
Error reading parameters file
Failed to read parameters file
19:51:28 (4132): called boinc_finish
</stderr_txt>
]]>
|
|
4)
Message boards :
News :
New Separation Runs Started
(Message 57691)
Posted 28 Mar 2013 by Senilix
Post: Are these WUs much longer than the previous ones? I'm asking because my BOINC client decided to switch to panic mode immediately after it got one of them. Here's the WU's XML definition:
<workunit>
<name>ps_separation_82_3s_dr8_2_1358941502_29559263</name>
<app_name>milkyway</app_name>
<version_num>100</version_num>
<rsc_fpops_est>89643799999999993000000000000000.000000</rsc_fpops_est>
<rsc_fpops_bound>89643799999999998000000000000000000.000000</rsc_fpops_bound>
<rsc_memory_bound>50000000.000000</rsc_memory_bound>
<rsc_disk_bound>15000000.000000</rsc_disk_bound>
<command_line>
-np 20 -p 0.943595320917666 23.0867525425274 -17.2245569434017 30.1253113951534 27.7632766885683 1.17406084491819 4.12454840047643 13.0079187392257 -11.5395608730614 39.1341196808498 22.0797624444123 0.980515889188198 1.62739609185373 12.7507249950431 13.1226277071983 41.4128890240099 30.2472704153508 6.16794836081251 3.20157651974716 5.24611199158244
</command_line>
<file_ref>
<file_name>p-82-3s-dr8-2.txt</file_name>
<open_name>astronomy_parameters.txt</open_name>
</file_ref>
<file_ref>
<file_name>stars-82-dr8.txt</file_name>
<open_name>stars.txt</open_name>
</file_ref>
</workunit>
<rsc_fpops_est> is much higher than it's used to be. BOINC deducted an estimated execution time of more then 24 hours... |
|
5)
Message boards :
Number crunching :
MilkyWay@Home v0.88 and MilkyWay@Home N-Body Simulation v0.66 (mt)
(Message 50409)
Posted 27 Jul 2011 by Senilix
Post: So it will ONLY got tasks from MW@Home, and not MW@Home N-Body Simulation (mt)? Exactly - and the MW@Home tasks will finish much faster because the optimized app will use the SSE3 instruction set of your INTEL PC. |
|
6)
Message boards :
Number crunching :
No cache limit for CPU app???
(Message 50400)
Posted 26 Jul 2011 by Senilix
Post: I'm running a GPU-only host on a dual core computer and my cache limit is set to 12 tasks - equaling 20 minutes. Now i noticed a wingman with a CPU-only host http://milkyway.cs.rpi.edu/milkyway/show_host_detail.php?hostid=302135 and a cache of more than 2700 tasks - equaling 2,000,000 minutes or more than 3 years???? Is there no cache limit in effect for the CPU app - or is this just a bug? |
|
7)
Message boards :
Number crunching :
MilkyWay@Home v0.88 and MilkyWay@Home N-Body Simulation v0.66 (mt)
(Message 50398)
Posted 26 Jul 2011 by Senilix
Post: Sorry, i didn't notice that there is no instruction.txt file contained in the download. First download the MilyWay@Home Windows 32/64 bit SSE3 INTEL CPUs ONLY Science App from http://www.mpits.net/opt_mw.php. This should be the correct one for your type of computer/OS. Installation of the optimized app is simple. 1) Stop BOINC service by going to the advanced menu and selecting "Shut down connected client, click okay and then click cancel. Then exit out of the BOINC Manager. 2) The zip file you've downloaded contains 2 files, an exe and an app_info.xml file. Extract both files into your BOINC\projects\milkyway.cs.rpi.edu_milkyway folder. Windows 98/SE/ME: C:\Windows\All Users\BOINC\ or C:\Windows\Profiles\All Users\BOINC\ Windows 2000/XP: C:\Documents and Settings\All Users\Application Data\BOINC\ Windows Vista and Windows 7: C:\ProgramData\BOINC\ 3) Restart BOINC service by reopening BOINC Manager. That's it, you're done! You can open your task manager to see that it's really the new optimized app crunching your WUs. BOINC will tell you that it found app_info.xml and is using an "anonymous platform". Happy crunching! |
|
8)
Message boards :
Number crunching :
MilkyWay@Home v0.88 and MilkyWay@Home N-Body Simulation v0.66 (mt)
(Message 50390)
Posted 26 Jul 2011 by Senilix
Post: Here at mw there is no way to select which type of CPU work your pc gets - at least not in your account configuration. But you can use an app_info.xml file to restrict your pc to wm@home apps only. The following message contains a download link to a SSE optimized exe, an appropriate app_info.xml and a textfile with instruction on how to install the exe and xml files. http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=2473&nowrap=true#50045 |
|
9)
Message boards :
Number crunching :
ps_separation and ps_test WUs all erroring out
(Message 49644)
Posted 27 Jun 2011 by Senilix
Post: All of the above mentioned WUs are erroring out on my ATI HD 5850. Same for my wingmen. As some of them are running nVidia cards, the issue seems to be WU related. |
|
10)
Message boards :
Number crunching :
Lastest Stock Apps - Optimized or Not
(Message 49613)
Posted 26 Jun 2011 by Senilix
Post: Here's a 'optimized' CPU app which was compiled using the Intel(R) C++ Compiler XE 12.0.4.196 for Windows. I am getting a 404 when i'm trying to download that opti app. |
|
11)
Message boards :
News :
Any remaining major credit or application problems?
(Message 49514)
Posted 23 Jun 2011 by Senilix
Post: Weekend would be better! +1 |
|
12)
Message boards :
News :
assimilator back up
(Message 45369)
Posted 31 Dec 2010 by Senilix
Post: Yes, thanks, i guess you are right. The credit is missing in BoincStats only, my Milkyway account is showing the correct amount. |
|
13)
Message boards :
News :
assimilator back up
(Message 45363)
Posted 31 Dec 2010 by Senilix
Post: Looks like there are more issues: I lost 200+K credits (which i may have earned due to the infamous assimilator/validator bug leading to duplicated credits for many WUs though) Downloads are failing now 31.12.2010 22:29:10 Milkyway@home Giving up on download of de_separation_16_3s_fix_1_253969_1293575410_search_parameters: file not found |
|
14)
Message boards :
Number crunching :
Validator stuck?
(Message 45062)
Posted 16 Dec 2010 by Senilix
Post: Workunits waiting for validation are piling up. And just to mention it: the issue with the assimilator doppelganger - with the effect of credits being double-counted - is still not solved. |
|
15)
Message boards :
News :
an update on the credit issue
(Message 44801)
Posted 7 Dec 2010 by Senilix
Post: Same here, am getting twice the usual credit for most WUs. But no need to rush into it, Travis ... you can fix thee issue between the years ... or next year ... or even ... :) |
|
16)
Message boards :
Number crunching :
super charged?
(Message 43841)
Posted 15 Nov 2010 by Senilix
Post: Double counting of wus seems to have been re-installed ... and i'm not complaining ;) |
|
17)
Message boards :
Number crunching :
Feeder & validator need kicking
(Message 38536)
Posted 10 Apr 2010 by Senilix
Post: Validator needs kicking again .... |
|
18)
Message boards :
Number crunching :
de_11_3s_free_5... computation error (cuda23)
(Message 37361)
Posted 15 Mar 2010 by Senilix
Post: I just noticed that a couple of my wu's ended with a computation error. I found that ALL of them have been of the de_11_3s_free_5 type whilst every wu I crunched successfully has NOT been of that type. Here's an example of the output of an erroneous wu.
The wu's are running up to ~99.9%, then stopping for a while and eventually ending with a computation error. I decided to abort any wu of that de_11_3s_free_5 type delivered to my box. EDIT: This seems to be CUDA-related only as my wingmen's ATI boxes are crunching them successfully. |
|
19)
Message boards :
Number crunching :
NVidia 196.75 WHQL withdrawn - Causes Overheating
(Message 37040)
Posted 8 Mar 2010 by Senilix
Post: EDIT: Sorry double post. The Collatz admin is recommending a fall back to version 196.21 but i found this in a Collatz thread from Zydor: Anyone using the 196.75 WHQL driver should roll back to 196.21 WHQL to avoid issues. NVidia have released a statement acknowledging the problem, withdrawn the 196.75 WHQL driver from its Site, and recommend rolling back to 196.21 WHQL (However see below). |
|
20)
Message boards :
Number crunching :
NVidia 196.75 WHQL withdrawn - Causes Overheating
(Message 37039)
Posted 8 Mar 2010 by Senilix
Post: The Collatz admin is recommending a fall back to version 196.21 but i found this in a Collatz thread from Zydor: Anyone using the 196.75 WHQL driver should roll back to 196.21 WHQL to avoid issues. NVidia have released a statement acknowledging the problem, withdrawn the 196.75 WHQL driver from its Site, and recommend rolling back to 196.21 WHQL (However see below). |
Next 20
©2024 Astroinformatics Group