Posts by Floyd

21) Message boards : Number crunching : What has your RAC done......since..... (Message 1375540)
Posted 2 Jun 2013 by Profile Floyd
Post:
I have 2 GPU WU's running at a time now , I had to take all caps out of the File name and out of the file it's self , The cpu use is still only 0.0261 instead of the 0.06 the config file had for it... I don't understand NOTHING anymore .. ( not that I ever did anyway )
22) Message boards : Number crunching : What has your RAC done......since..... (Message 1375528)
Posted 2 Jun 2013 by Profile Floyd
Post:
Looks okay - save it in the SETI project directory. BOINC mgr. has a menu option on the advanced tab to "read config file", so you don't even have to restart BOINC.


Ok I saved it in the Seti Project dir... It shows to be here :

C:\ProgramData\BOINC\projects\setiathome.berkeley.edu\App_config.xml

Still only running 1 GPU task at a time...
What did I mess up this time...? :{

6/2/2013 1:23:25 PM | | No config file found - using defaults
6/2/2013 1:23:25 PM | | Starting BOINC client version 7.0.64 for windows_x86_64
6/2/2013 1:23:25 PM | | log flags: file_xfer, sched_ops, task
6/2/2013 1:23:25 PM | | Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
6/2/2013 1:23:25 PM | | Data directory: C:\ProgramData\BOINC
6/2/2013 1:23:25 PM | | Running under account Floyds

I did the read config and restarted Boing also.

I will take the CAP A off the app_config and then retry it , maybe that is the problem.


Well that allowed it to find the file , but !!!

6/2/2013 2:04:56 PM | | No config file found - using defaults
6/2/2013 2:04:56 PM | | Starting BOINC client version 7.0.64 for windows_x86_64
6/2/2013 2:04:56 PM | | log flags: file_xfer, sched_ops, task
6/2/2013 2:04:56 PM | | Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
6/2/2013 2:04:56 PM | | Data directory: C:\ProgramData\BOINC
6/2/2013 2:04:56 PM | | Running under account Floyds
6/2/2013 2:04:56 PM | | Processor: 4 AuthenticAMD AMD Phenom(tm) II X4 925 Processor [Family 16 Model 4 Stepping 3]
6/2/2013 2:04:56 PM | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 htt pni cx16 popcnt syscall nx lm svm sse4a osvw ibs skinit wdt page1gb rdtscp 3dnowext 3dnow
6/2/2013 2:04:56 PM | | OS: Microsoft Windows 7: Professional x64 Edition, Service Pack 1, (06.01.7601.00)
6/2/2013 2:04:56 PM | | Memory: 3.92 GB physical, 10.75 GB virtual
6/2/2013 2:04:56 PM | | Disk: 1.36 TB total, 432.46 GB free
6/2/2013 2:04:56 PM | | Local time is UTC -5 hours
6/2/2013 2:04:56 PM | | CUDA: NVIDIA GPU 0: GeForce GTX 460 SE (driver version 314.22, CUDA version 5.0, compute capability 2.1, 1024MB, 939MB available, 922 GFLOPS peak)
6/2/2013 2:04:56 PM | | OpenCL: NVIDIA GPU 0: GeForce GTX 460 SE (driver version 314.22, device version OpenCL 1.1 CUDA, 1024MB, 939MB available, 922 GFLOPS peak)
6/2/2013 2:04:56 PM | SETI@home | Found app_config.xml6/2/2013 2:04:56 PM | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 5953062; resource share 100
6/2/2013 2:04:56 PM | SETI@home | General prefs: from SETI@home (last modified 17-May-2013 09:48:48)
6/2/2013 2:04:56 PM | SETI@home | Computer location: work
6/2/2013 2:04:56 PM | | General prefs: using separate prefs for work
6/2/2013 2:04:56 PM | | Reading preferences override file
6/2/2013 2:04:56 PM | | Preferences:
6/2/2013 2:04:56 PM | | max memory usage when active: 2005.78MB
6/2/2013 2:04:56 PM | | max memory usage when idle: 3610.40MB
6/2/2013 2:04:56 PM | | max disk usage: 10.00GB
6/2/2013 2:04:56 PM | | (to change preferences, visit a project web site or select Preferences in the Manager)
6/2/2013 2:04:56 PM | | Not using a proxy

still only running 1 gpu wu at a time , i'm still not doing something right...!
23) Message boards : Number crunching : What has your RAC done......since..... (Message 1375509)
Posted 2 Jun 2013 by Profile Floyd
Post:
Looks okay - save it in the SETI project directory. BOINC mgr. has a menu option on the advanced tab to "read config file", so you don't even have to restart BOINC.


Ok I saved it in the Seti Project dir... It shows to be here :

C:\ProgramData\BOINC\projects\setiathome.berkeley.edu\App_config.xml

Still only running 1 GPU task at a time...
What did I mess up this time...? :{

6/2/2013 1:23:25 PM | | No config file found - using defaults
6/2/2013 1:23:25 PM | | Starting BOINC client version 7.0.64 for windows_x86_64
6/2/2013 1:23:25 PM | | log flags: file_xfer, sched_ops, task
6/2/2013 1:23:25 PM | | Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
6/2/2013 1:23:25 PM | | Data directory: C:\ProgramData\BOINC
6/2/2013 1:23:25 PM | | Running under account Floyds

I did the read config and restarted Boing also.

I will take the CAP A off the app_config and then retry it , maybe that is the problem.
24) Message boards : Number crunching : What has your RAC done......since..... (Message 1375500)
Posted 2 Jun 2013 by Profile Floyd
Post:
The app_config file you have here , would it work ( as it is ) on a quad core CPU and a GTX 460 SE GPU ? I'm not running any AP's at all Just V7 WU's ....

It would work, but I'd make a couple of changes:
in the v7 section:
<cpu_usage>0.5</cpu_usage>
I'd reduce the value to .06, you don't need to reserve that much cpu time.

I'd also omit the v6 section entirely unless you want to crunch the resends. Probably won't get any at all after the installer becomes available.

I'd keep the astropulse section in there in case you want to run it someday.
<max_concurrent>6</max_concurrent>
I'd set the limit lower than 6, I use 2.

The gpu usage value of .5 will give you 2 at a time which is probably right for your gpu.



Ok , I left the V6 in case we get any resends , won't hurt any I guess ?


does this look right ?

<App_config>
<app>
<name>setiathome_v7</name>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.06</cpu_usage>
</gpu_versions>
</app>

<app>
<name>setiathome_enhanced</name>
<max_concurrent>4</max_concurrent>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.06</cpu_usage>
</gpu_versions>
</app>

<app>
<name>astropulse_v6</name>
<max_concurrent>2</max_concurrent>
<gpu_versions>
<gpu_usage>1</gpu_usage>
<cpu_usage>1.0</cpu_usage>
</gpu_versions>
</app>

</App_config>
25) Message boards : Number crunching : What has your RAC done......since..... (Message 1375482)
Posted 2 Jun 2013 by Profile Floyd
Post:
MB v7 tasks are being assigned shockingly low credit scores (like 30%-50% of what v6 was giving on average).

Yes, if was programmed to give less rac for reasons I can't comprehend.

Also my Astropulse nvidia credit is one fifth when my partner is an ATI GPU!!!

To run more the one app at a time on you GPU, you will need to creat an app_config.xml file:


<app_config>
<app>
<name>setiathome_v7</name>
<gpu_versions>
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.5</cpu_usage>
</gpu_versions>
</app>

<app>
<name>setiathome_enhanced</name> 
<max_concurrent>6</max_concurrent> 
<gpu_versions> 
<gpu_usage>0.5</gpu_usage> 
<cpu_usage>0.5</cpu_usage> 
</gpu_versions> 
</app> 

<app>
<name>astropulse_v6</name>
<max_concurrent>6</max_concurrent>
<gpu_versions>
<gpu_usage>.50</gpu_usage>
<cpu_usage>1.0</cpu_usage>
</gpu_versions>
</app>

</app_config>


The app_config file you have here , would it work ( as it is ) on a quad core CPU and a GTX 460 SE GPU ? I'm not running any AP's at all Just V7 WU's ....
26) Message boards : Number crunching : Boinc guidance needed (Message 1373897)
Posted 30 May 2013 by Profile Floyd
Post:
Normally when you uninstall and then re-install you have to tell Boinc what projects you want to run , from the projects page . also when you reattach to a project you have to sign in so Boinc will be able to find your old account to reconnect to. otherwise it might think it is a different computer and start a new account.

Not true. That would only apply if:

* You manually deleted the (normally hidden) data directory.
* You deliberately chose a non-default data directory location for either the old or new installation, and failed to align them (though the BOINC installer normally remembers that for you, unless you clean the registry)
* You tried to install a 64-bit BOINC after uninstalling a 32-bit BOINC (or vice versa), without checking data directory locations as above.

The key to all project attachments and account settings is the data directory. Lose that, and you lose everything. Let BOINC manage it automatically, and everything is retained.


Ok , after I uninstall a program , I use CCleaner to do a light cleaning of the Registry to remove any old info so it doesn't corrupt on re-install , Kind of when you remove any drivers to do a clean install.

If he had done that also , then he would have to re-attach , was what I was thinking, and sign-in again... That was all I could think of at the time...
Not the first time I have messed up , guess I should just let the more experienced people help him... Sorry...
27) Message boards : Number crunching : Boinc guidance needed (Message 1373860)
Posted 30 May 2013 by Profile Floyd
Post:
No, but Seti appears on Boinc Manager Project tab.

How does one reattach?



Normally when you uninstall and then re-install you have to tell Boinc what projects you want to run , from the projects page . also when you reattach to a project you have to sign in so Boinc will be able to find your old account to reconnect to. otherwise it might think it is a different computer and start a new account.

According to what the computers page shows for you , that PC is running WU's , could you have the program running in the background even when you close the boinc manager window ?
28) Message boards : Number crunching : Boinc guidance needed (Message 1373831)
Posted 30 May 2013 by Profile Floyd
Post:
Yes



OK , Did you re-attach to the seta project when you re-installed Boinc ?
Then re-sign in to your account for Seti at home , so it will recognize the PC as the same one ?
29) Message boards : Number crunching : Boinc guidance needed (Message 1373823)
Posted 30 May 2013 by Profile Floyd
Post:
Per Seti, I have 85 tasks ready to run.

No tasks appear in Boinc Manager.

No tasks are running.

Update gets this sequence:

67 SETI@home 5/30/2013 5:02:19 PM update requested by user
68 SETI@home 5/30/2013 5:02:20 PM Sending scheduler request: Requested by user.
69 SETI@home 5/30/2013 5:02:20 PM Not reporting or requesting tasks
70 SETI@home 5/30/2013 5:02:21 PM Scheduler request completed

I have uninstalled Boinc and reinstalled with reboot in between.

I'd appreciate any suggestions.



Is this on : All tasks for computer 6833184
http://setiathome.berkeley.edu/results.php?hostid=6833184
Is this the Computer you are having problems on ?
30) Message boards : News : AVG 2013 virus scanner false positive on SETI@home 7 for Windows (Message 1373814)
Posted 30 May 2013 by Profile Floyd
Post:
If you're willing, could you please download http://boinc2.ssl.berkeley.edu/beta/download/setiathome_7.00_windows_intelx86_unpacked.exe
and scan it with your virus scanner to see if it reports a problem with the uncompressed file?

And if that checks out, try to scan a recompressed version, just to be sure. http://boinc2.ssl.berkeley.edu/beta/download/setiathome_7.00_windows_intelx86_repacked.exe


Sir:
I downloaded both and AVG 2013 scanned them in the download folder and found nothing.

EDIT:
That was after it found them and I told it to allow , so it was added to the exception list in avg , Just remembered that.
31) Message boards : Number crunching : Now that v7 has rolled out.......... (Message 1373781)
Posted 30 May 2013 by Profile Floyd
Post:
I just noticed something that I don't remember seeing on the V6 WU's for GPU.

On PC ( #1 ) 5953062 the GPU wu is showing " Running ( 0.0236 CPU's + 1 NVidia GPU )" CUDA 32 WU

On PC ( #2 ) 6994641 the GPU wu is showing " Running ( 0.00579 CPU's + 1 NVidia GPU )" CUDA 42 WU

The difference in CPU usage... Can this be because of the different Video cards ?

#1 has the GTX 460SE and #2 has the GT 430 .

OR
because of the different Cuda apps being used at this time ?
32) Message boards : Number crunching : Now that v7 has rolled out.......... (Message 1373743)
Posted 30 May 2013 by Profile Floyd
Post:
Everything seems to be good , Just REAL long crunching estimates ( compared to before) , what used to take ( CPU ) 1.5-2 hrs is now showing 4-5 hours ,
GPU is now over 30 min long .

That sound about right ?

That sounds about right, the 7.00 app includes support for AVX, and now uses the latest FFTW 3.3 dll too, which also gives a further speedup compared to the old FFTW 3.1.1 dll,
Hosts with AVX support will get a big speedup, my i7-2600K @4.7GHz does Stock 7.00 CPU Wu's in ~2 to 2½ Hrs, while my C2D T8100 @2.1GHz is doing them in 6 Hrs.

Claggy



If the V7 GPU app's are doing more Intensive work on the Wu's , is it maybe more needed to free up 1 core on cpu , to be more effective for GPU work ?
With the V6 apps , I was getting about the same results with one core free as I was without one free, as far as credit and time went anyway.
33) Message boards : Number crunching : Now that v7 has rolled out.......... (Message 1373669)
Posted 30 May 2013 by Profile Floyd
Post:
An estimate is just an estimate, run a few tasks for real and see what happens. But they will take longer than before, yes. That's the whole purpose of this rollout - to study the WUs in greater detail. And more work == more time.


That's ok... I LIKE it... LOL... I was just putting out the info for comparative purposes ,(for the GTX 460SE anyway ).
34) Message boards : Number crunching : Now that v7 has rolled out.......... (Message 1373652)
Posted 30 May 2013 by Profile Floyd
Post:
OK... I have set up and am doing V7 WU's now , here is an excerpt from the event log:

5/30/2013 12:38:27 PM | | Starting BOINC client version 7.0.64 for windows_x86_64
5/30/2013 12:38:27 PM | | log flags: file_xfer, sched_ops, task
5/30/2013 12:38:27 PM | | Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
5/30/2013 12:38:27 PM | | Data directory: C:\ProgramData\BOINC
5/30/2013 12:38:27 PM | | Running under account Floyds
5/30/2013 12:38:27 PM | | Processor: 4 AuthenticAMD AMD Phenom(tm) II X4 925 Processor [Family 16 Model 4 Stepping 3]
5/30/2013 12:38:27 PM | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 htt pni cx16 popcnt syscall nx lm svm sse4a osvw ibs skinit wdt page1gb rdtscp 3dnowext 3dnow
5/30/2013 12:38:27 PM | | OS: Microsoft Windows 7: Professional x64 Edition, Service Pack 1, (06.01.7601.00)
5/30/2013 12:38:27 PM | | Memory: 3.92 GB physical, 10.75 GB virtual
5/30/2013 12:38:27 PM | | Disk: 1.36 TB total, 433.24 GB free

5/30/2013 12:38:27 PM | | Local time is UTC -5 hours
5/30/2013 12:38:27 PM | | CUDA: NVIDIA GPU 0: GeForce GTX 460 SE (driver version 314.22, CUDA version 5.0, compute capability 2.1, 1024MB, 951MB available, 829 GFLOPS peak)
5/30/2013 12:38:27 PM | | OpenCL: NVIDIA GPU 0: GeForce GTX 460 SE (driver version 314.22, device version OpenCL 1.1 CUDA, 1024MB, 951MB available, 829 GFLOPS peak)
5/30/2013 12:38:27 PM | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 5953062; resource share 100
5/30/2013 12:38:27 PM | SETI@home | General prefs: from SETI@home (last modified 17-May-2013 09:48:48)

Everything seems to be good , Just REAL long crunching estimates ( compared to before) , what used to take ( CPU ) 1.5-2 hrs is now showing 4-5 hours ,
GPU is now over 30 min long .

That sound about right ?
35) Message boards : Number crunching : Now that v7 has rolled out.......... (Message 1373145)
Posted 29 May 2013 by Profile Floyd
Post:
V7 is showing up on application page now !
36) Message boards : Number crunching : Now that v7 has rolled out.......... (Message 1372761)
Posted 28 May 2013 by Profile Floyd
Post:
Hey...
Is there any new info on the V7 release , as far as what us "Tech challenged"
crunchers should do to get ready for the V7 situation ?

Maybe links to what and where to get what we will need to keep up with the newer
programs ?

Is it still coming out tomorrow after todays downtime ( Tuesdays outage) ?

Inquiring ( Small and feeble) Minds want to know. ;-)


KISS? I can try...

If you are on stock -> do nothing, except to doublecheck that v7 is ticked in your preferences.

If you run optimised seti only on GPU and run boinc 7.0.64 -> go stock: run down your v6 cache, when empty remove app_info.xml, tick v7 and untick v6, reset project to get stock apps. Use app_config.xml to run several tasks at once.
[you won't get v6 resends, but you would be processing them with 6.08/6.09/6.10 which isn't really desirable (if you have an NV that is)]

If you are optimised and run CPU and GPU you can

a) go stock (as above) and patiently wait until such time as a Lunatics installer becomes available [ETA Monday after launch, provided I don't suffer a nervous breakdown before that] and then decide whether you want to optimise again NB only CPU apps are optimised at this point since the GPU apps delivered as stock are actually the optimised ones.

b) wait paitently until the installer becomes available and hope there are enough v6 resends to keep you running, ignoring the protests of boinc about you not having a v7 app in your app_info.xml

c) cobble together your own v7 entries. NB AKv8b2 is NOT v7 compatible. you could upgrade your NV/ATI parts though. And patiently wait for the installer to become available to get the optimised CPU apps.

d) panic and demand an installer to be availbale from yesterday on, causing me to have a hissy fit and thus delaying installer availablity until somebody who has very limited time has gone 'there, there'.

I suggest sticking to options a) to c)

Installer availability will be announced through the usual channels.




LOL... I'm not sure where the --- "KISS? I can try..."
Part came in from , unless I did a serious typo... LOL
any way , I do have the opti apps on both CPU and GPU , so I guess I'll just leave the V7 unchecked until the apps become available . about Monday or so...and I don't demand anything from anybody ... LOL... so you won't have to worry about ME griping or complaining to you or about you either...
I appreciate all that you and the others who have the skills and knowledge to do what you do , do it on YOUR Free time , to help all of us do what we do. !!!

Thanks ( from me at Least ) for all your ( and Others ) effort to make things work so that we can work !

I was just trying to see if I missed something or needed to do something else to be ready for the changes coming.
Thanks again...
37) Message boards : Number crunching : Now that v7 has rolled out.......... (Message 1372741)
Posted 28 May 2013 by Profile Floyd
Post:
Hey...
Is there any new info on the V7 release , as far as what us "Tech challenged"
crunchers should do to get ready for the V7 situation ?

Maybe links to what and where to get what we will need to keep up with the newer
programs ?

Is it still coming out tomorrow after todays downtime ( Tuesdays outage) ?

Inquiring ( Small and feeble) Minds want to know. ;-)
38) Message boards : Number crunching : Now that v7 has rolled out.......... (Message 1371644)
Posted 25 May 2013 by Profile Floyd
Post:
Well... ( deep subject for such a shallow mind as mine ! )

I think I will just uncheck V7 Wu's in Boinc / or / Prefs ,until all this gets figured out , if that is even possible !
39) Message boards : Number crunching : ASUS MB Owners (Message 1370108)
Posted 20 May 2013 by Profile Floyd
Post:
Not being able to install an OS is a pretty critical item. For most people that would be a show stopper. I doubt a weak Power Supply would cause that problem, it could, but it would have to be so weak that running SETI at all would be nearly impossible. That doesn't appear to be the case. You might want to concentrate on asking Corsair about those BSODs when running the install disk. Those BSODs point to a ram problem, no matter how many times it passes that test. My Intel board passed that test too, it was still the ram, the same ram that runs perfectly fine on my ECS board. Try testing the install disk again. To be safe you would need a different HD, I would remove the working drive you have and test the installer on a different drive. Until you are able to run the Installer successfully, there isn't any point in proceeding with anything else, IMO. Like I said, not being able to install the OS would be a Fatal event in most cases.


When I get able to , I plan on getting a new HD , a SATA 3 version ,Then do a clean install of the OS , and if that works I'll use that for the Boot sys , and keep this older drive to be used as a back up drive.
This new ASUS MB was built for the newer drives , that may be part of the problem also.

I'm just hoping for the best and grasping at straws right now !

As far as I know, all SATA controllers are backwards compatible, Backward and forward compatibility. Considering you've already exchanged the drive, the chances of the HD causing the problem is extremely remote. You can always call ASUS and ask their advice. Make sure you tell them about the problems with the Win 7 Install disk. As a side note, neither of my machines that passed the ram tests, but still had ram problems, gave any problems with the OS installer. The problem with the OS installer is a huge red flag.


Take a look at the specs of this MB:

http://www.asus.com/us/Motherboards/M5A97_R20/#overview

This OS was installed on a HD in my old ECS MB ( MCP61M-M3 ) , When I changed the MB , I just tried to install the newer drivers and all on the same HD and OS , Which may be causing some sort of conflict with this new MB .
There is too much data on my 1.5 TB HD that I don't want to lose.
That's why I said a new HD and do a clean install might correct these problems.

That's what I am hoping anyway !
40) Message boards : Number crunching : ASUS MB Owners (Message 1370080)
Posted 20 May 2013 by Profile Floyd
Post:
I had another computation error on this WU.

http://setiathome.berkeley.edu/workunit.php?wuid=1249524318

I was suspending Work several times trying to get boinc tasks set up and I read somewhere on here that that will sometimes cause a compute error .

I will wait for a while and see if it repeats ,I hope not...!

SETI@home 6.10 setiathome_enhanced (cuda_fermi) 26se12ad.21118.25016.3.11.188_0 00:15:24 (00:02:26) 5/20/2013 12:25:47 PM 5/20/2013 12:32:13 PM 0.04C + 1NV 15.80 Reported: Computation error (194,) Floyd


Previous 20 · Next 20


 
©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.