A pretty minor update. Report problems here.

I added a 32 bit OS X one, although it doesn't get multithreaded.

Update: For Windows, I got the names of the DLLs backwards for BOINC the first time for 0.60. Then BOINC's copying of the DLLs wasn't working for no particular reason. Then it turns out one of the DLLs I uploaded somehow was corrupted, but in any case it seems to now be working. So the Windows ones now say version 0.66, but nothing actually changed from the earlier ones in case you noticed those.

by any error in my house!

this may be due to my file app_info?

<app_info>
<app>
<name>milkyway</name>
</app>
<file_info>
<name>milkyway_separation_0.82_windows_x86_64__cuda_opencl.exe</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway</app_name>
<version_num>82</version_num>
<flops>1.0e11</flops>
<avg_ncpus>0.05</avg_ncpus>
<max_ncpus>1</max_ncpus>
<plan_class>cuda</plan_class>
<coproc>
<type>CUDA</type>
<count>1</count>
</coproc>
<cmdline></cmdline>
<file_ref>
<file_name>milkyway_separation_0.82_windows_x86_64__cuda_opencl.exe</file_name>
<main_program/>
</file_ref>
</app_version>
<app>
<name>milkyway_nbody</name>
<user_friendly_name>MilkyWay@Home nbody Simulation</user_friendly_name>
</app>
<file_info>
<name>milkyway_nbody_0.66_windows_x86_64__mt.exe</name>
<executable/>
</file_info>
<file_info>
<name>libgomp_nbody_0.66.dll</name>
<executable/>
</file_info>
<file_info>
<name>pthreadGC2_nbody_0.66.dll</name>
<executable/>
</file_info>
<app_version>
<app_name>milkyway_nbody</app_name>
<version_num>66</version_num>
<plan_class>mt</plan_class>
<avg_ncpus>2</avg_ncpus>
<max_ncpus>2</max_ncpus>
<cmdline>--nthreads=2</cmdline>
<file_ref>
<file_name>milkyway_nbody_0.66_windows_x86_64__mt.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>libgomp_nbody_0.66.dll</file_name>
</file_ref>
<file_ref>
<file_name>pthreadGC2_nbody_0.66.dll</file_name>
</file_ref>
</app_version>
</app_info>

---

Team Alliance francophone, boinc: 7.0.18

GA-P55-UD5, i7 860, Win 7 64 bits, 8g DDR3, GTX 470

You have 2 problems. First you're trying to use the 32 bit DLLs with the 64 bit application. Second, you're missing the stuff to rename the DLLs. You need to add the blah and

It should be something like this (though I haven't tried it)


libgomp_64_nbody_0.66.dll
libgomp_64-1.dll





pthreadGC2_64_nbody_0.66.dll
pthreadGC2_64.dll

thanks you

---

Team Alliance francophone, boinc: 7.0.18

GA-P55-UD5, i7 860, Win 7 64 bits, 8g DDR3, GTX 470

Can linux users just stick with nbody 0.40? You just did this previous version in March.

Thanks Matt. I now have an Nbody WU that has made it to 18:50 run time, 34% progress, without crashing. I haven't had one go this far for weeks.

---

Have had 1 WU that completed & validated after weeks of 'computation errors'.

Thank-you.

Can linux users just stick with nbody 0.40? You just did this previous version in March.

0.40 is fine if it's working for you. The only major change isn't actually used right now. There was one crash which was much more likely on Windows that I fixed that I think has a small chance of happening on Linux (Particularly 32 bit) but I'm not sure

Thank You !!!!!

I have had all of these on my WinXP64 systems fail since the last version.

They are now going through.

Thanks !

2 Nbody completed and validated. Life is good.

---

Another with no credit. I'm going to abort all of these tasks untill this is fixed.

Task 56429160
Name de_nbody_orphan_test_2model_4_43195_1308787963_0
Workunit 38642788
Created 23 Jun 2011 | 0:12:47 UTC
Sent 23 Jun 2011 | 0:30:32 UTC
Received 23 Jun 2011 | 13:17:22 UTC
Server state Over
Outcome Success
Client state Done
Exit status 0 (0x0)
Computer ID 1500
Report deadline 5 Jul 2011 | 0:30:32 UTC
Run time 0.00
CPU time 9,552.84
Validate state Valid
Credit 0.00
Application version MilkyWay@Home N-Body Simulation v0.66 (mt)

Stderr output
<core_client_version>5.10.45</core_client_version>
<![CDATA[
<stderr_txt>
<search_likelihood>-1415.34809342267</search_likelihood>
<search_application>milkywayathome nbody 0.62 Windows x86 double OpenMP Crlibm</search_application>
09:05:24 (1628): called boinc_finish

</stderr_txt>
]]>

---

Doesn't expecting the unexpected make the unexpected the expected?
If it makes sense, DON'T do it.

5.10.45

This might be your problem. Try updating your boinc.

Another with no credit. I'm going to abort all of these tasks untill this is fixed.

Task 56429160
Name de_nbody_orphan_test_2model_4_43195_1308787963_0
Workunit 38642788
Created 23 Jun 2011 | 0:12:47 UTC
Sent 23 Jun 2011 | 0:30:32 UTC
Received 23 Jun 2011 | 13:17:22 UTC
Server state Over
Outcome Success
Client state Done
Exit status 0 (0x0)
Computer ID 1500
Report deadline 5 Jul 2011 | 0:30:32 UTC
Run time 0.00
CPU time 9,552.84
Validate state Valid
Credit 0.00
Application version MilkyWay@Home N-Body Simulation v0.66 (mt)

Stderr output
5.10.45

-1415.34809342267
milkywayathome nbody 0.62 Windows x86 double OpenMP Crlibm
09:05:24 (1628): called boinc_finish


]]>

Im looking into this. It's kind of weird why it's awarding zero credit. Could it because you're running an older boinc client?

---

This indeed is a problem with the older clients not playing nice with BOINC's new credit policy. I'm going to see if I can do some tweaking to make it better, but if possible you should probably upgrade your client.

---

Did some updates to the credit policy. The old BOINC clients should be awarded a more sane amount of credit (ie. not 0).

---