Astropulse FAQ

Message boards : Number crunching : Astropulse FAQ
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 9 · 10 · 11 · 12 · 13 · 14 · Next

AuthorMessage
Profile Jug Head
Avatar

Send message
Joined: 21 Oct 00
Posts: 6
Credit: 93,466
RAC: 0
United States
Message 868439 - Posted: 23 Feb 2009, 7:30:06 UTC

I have an old machine. Is there anyway I can NOT get Astopulse WUs? They are way to long vs the 5 to 6 hour version?
ID: 868439 · Report as offensive
Profile Byron S Goodgame
Volunteer tester
Avatar

Send message
Joined: 16 Jan 06
Posts: 1145
Credit: 3,936,993
RAC: 0
United States
Message 868443 - Posted: 23 Feb 2009, 7:43:19 UTC - in response to Message 868439.  
Last modified: 23 Feb 2009, 7:47:41 UTC

I have an old machine. Is there anyway I can NOT get Astopulse WUs? They are way to long vs the 5 to 6 hour version?

Hi,

In your SETI@home preferences page, under "Run only the selected applications" uncheck "Astropulse", and uncheck "If no work for selected applications is available, accept work from other applications?". then clcik update preferences.
ID: 868443 · Report as offensive
_hiVe*

Send message
Joined: 7 Aug 04
Posts: 9
Credit: 4,980,874
RAC: 0
Slovakia
Message 868454 - Posted: 23 Feb 2009, 9:24:20 UTC

Hello everyone~

I've recently started using the latest optimized AP app's. (v5.03v112)
It made me wonder though, how often will I have to check whether there has been an app. revision / update ?

Can you roughly estimate (based on previous experiences), how often are the updated stock app's released ?

Also, what will happen if I don't use the latest optimized AP app's, will my WU's error out?

I've done a short search, but I couldn't dig out anything specific that would answer my questions.
Thanks so much for the answers~

ID: 868454 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14679
Credit: 200,643,578
RAC: 874
United Kingdom
Message 868469 - Posted: 23 Feb 2009, 10:07:50 UTC

Two very good questions, worthy of any FAQ.

Astropulse was released in August 2008 (v4.35). There have been two major revisions since then: v5.00 in November 2008, and v5.03 in February 2009. So simplistically, and going by a pretty brief historical precedent, you might say "every three months". The other main SETI app is more mature, and is updated less frequently, so the frequency might go down in the future: or a major donor might come forward, enable them to hire more specialist staff, and somebody might concentrate on developing the code more quickly. All we can say is that new releases will come when they come, and probably when we least expect them (though you will get at least an idea of what's in the pipeline if you join the Beta project).

What happens if you don't use the latest optimised application? That depends how the Berkeley team handles the transition.

In November, with the transition from v4.35 to v5.00, you would have noticed nothing - like the anonymous user who crunched task 1165185917 a few days ago. That work will never validate: it will get no credit, and it will never contribute to the science of Astropulse. It's a complete waste of electricity, with no merit at all. You need to keep an eye on the message boards to avoid that happening to you.

The February transition, on the other hand, kept the old and the new work entirely separate. So if anyone fails to upgrade to the latest version, their supply of astropulse tasks will simply dry up (gradually - it takes time to work the last few resends out of the system).

Whichever way they do the next transition, it's unlikely that tasks will error out, and you certainly won't see any messages about it in BOINC Manager. There may be a news item on the front page, but it's unlikely - neither of the last two versions have been announced there (and hence you won't find out by RSS). The message boards are probably the only reliable place for finding out what's going on.
ID: 868469 · Report as offensive
dyeman
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 23
Credit: 8,493,140
RAC: 0
Australia
Message 868478 - Posted: 23 Feb 2009, 10:35:34 UTC - in response to Message 868454.  

One way to keep tabs on new apps is to subscribe to the lunatics forum that announces new optimized apps. You should then be notified whenever any new versions are released.

http://lunatics.kwsn.net/discussion-forum/releases-all-released-versions-are-announced-here.0.html

ID: 868478 · Report as offensive
_hiVe*

Send message
Joined: 7 Aug 04
Posts: 9
Credit: 4,980,874
RAC: 0
Slovakia
Message 868488 - Posted: 23 Feb 2009, 11:50:47 UTC

Thank you both, it certainly cleared up a few queries on my part and gave me a better insight overall ;)

Have a pleasant day~
ID: 868488 · Report as offensive
Profile S@NL - Eesger - www.knoop.nl
Avatar

Send message
Joined: 7 Oct 01
Posts: 385
Credit: 50,200,038
RAC: 0
Netherlands
Message 882382 - Posted: 4 Apr 2009, 23:18:50 UTC
Last modified: 4 Apr 2009, 23:19:06 UTC

It is probably old news to you all, but since I have started with AP (mainly hoping to keep crunching for SETI due to Thumper) my RAC has dropped about 1k and my pending is about 3.5 times to what it was (and I didn't even start about my backupproject, CPDN, whoom will probably thank you for your problems ;) )

All could have been expected, but it is new to me.

One thought though, when an std. wu (MB) is deemed invalid I'dd loose about 40 credits.. (bummer, but no disaster), but if somehow an AP-wu is deemed invalid I'dd lose 1.2k credits (big bummer ;) ) Is it a valid question to ask for more "security" to (in)validate a AP-wu? (say let 'm run three times for validation?) or??

just a thought..
The SETI@Home Gauntlet 2012 april 16 - 30| info / chat | STATS
ID: 882382 · Report as offensive
OzzFan Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Apr 02
Posts: 15691
Credit: 84,761,841
RAC: 28
United States
Message 882391 - Posted: 4 Apr 2009, 23:40:40 UTC - in response to Message 882382.  

It is probably old news to you all, but since I have started with AP (mainly hoping to keep crunching for SETI due to Thumper) my RAC has dropped about 1k and my pending is about 3.5 times to what it was (and I didn't even start about my backupproject, CPDN, whoom will probably thank you for your problems ;) )

All could have been expected, but it is new to me.


See my post here explaining why your RAC drops when doing AP work.

One thought though, when an std. wu (MB) is deemed invalid I'dd loose about 40 credits.. (bummer, but no disaster), but if somehow an AP-wu is deemed invalid I'dd lose 1.2k credits (big bummer ;) ) Is it a valid question to ask for more "security" to (in)validate a AP-wu? (say let 'm run three times for validation?) or??

just a thought..


When I went to the doctor and told him it hurts when I touch my elbow, he told me to stop touching my elbow. :)

As long as our computers are running fine, the majority of our workunits returned will validate. Sure, it sucks when an AP doesn't validate, but that's the risk we all take when crunching. A quorum of 3 wouldn't change this, because if our result wasn't accurate enough, the other two will get credit while we still get zero.
ID: 882391 · Report as offensive
rdo

Send message
Joined: 30 Nov 00
Posts: 1
Credit: 40,126
RAC: 0
United States
Message 883140 - Posted: 7 Apr 2009, 21:29:01 UTC - in response to Message 882391.  
Last modified: 7 Apr 2009, 21:43:32 UTC

is there something wrong with astropulse? the avg turnaround time jumped up 9 hours while the seti average droped 10 hours over the time of the server update.

Result turnaround time (last hour average) 76.48 hours 170.14 hours 3h
Result turnaround time (last hour average) 66.22 hours 179.67 hours 0m

and the 06fe09aa file looks half grey/error. i was thinking it was my pc but i should be able to compete the ap in less than the agerage time. i have a dual core and aps running on both cores, one keeps droping the completion % about .300 - .400 ever other % or so. it's taking me an hour to do .650 so thats quite a time waste every time. the other core is working slower and started 3 hours earlier. the core that keeps droping the completed % will catch up to about 1% behind then drop to about 1.6% behind.

and one of my wingmen was finished with his task before i was at 40%. he is running optamised, but he shouldn't be that much faster. his slower pc did in 17h 15m what will take me almost 150 hours to do. i have a faster processor, his fps is 2363.3 while mine is at 2714.07. how is he 3-4X faster? is the stock app really that unoptamized? http://setiathome.berkeley.edu/workunit.php?wuid=429575679

i'm going to finish this work unit but have already unckecked the astropulse boxes so i dont get anymore in the future. it's going to take me about 150 hours to do one, slightly faster than average. the regular seti my pc cranks out in less than 1/3 of average. it might be that ap dosn't like amd.
ID: 883140 · Report as offensive
Dotsch
Volunteer tester
Avatar

Send message
Joined: 9 Jun 99
Posts: 2422
Credit: 919,393
RAC: 0
Germany
Message 884928 - Posted: 13 Apr 2009, 14:09:45 UTC

Is Astropulse 5.00 still a vaild application, or was it complete obsolated by the release of the Astropulse V_5 (5.03) ?
ID: 884928 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14679
Credit: 200,643,578
RAC: 874
United Kingdom
Message 884936 - Posted: 13 Apr 2009, 14:30:28 UTC - in response to Message 884928.  

Is Astropulse 5.00 still a vaild application, or was it complete obsolated by the release of the Astropulse V_5 (5.03) ?

Astropulse v5.00 would have been needed to complete any remaining unvalidated WUs in the database, but I think they've probably all gone now - I haven't seen one for ages. So I think you can safely treat it as obsolete - no work has been split in a way that requires it for over 2 months (12 February), and none ever will be again.
ID: 884936 · Report as offensive
Urs Echternacht
Volunteer tester
Avatar

Send message
Joined: 15 May 99
Posts: 692
Credit: 135,197,781
RAC: 211
Germany
Message 884951 - Posted: 13 Apr 2009, 15:50:09 UTC - in response to Message 884936.  

Is Astropulse 5.00 still a vaild application, or was it complete obsolated by the release of the Astropulse V_5 (5.03) ?

Astropulse v5.00 would have been needed to complete any remaining unvalidated WUs in the database, but I think they've probably all gone now - I haven't seen one for ages. ...

I would be careful with that assumption. Last week i've completed at least one of these Astropulse v5.00 wus. There might still be "a hand full" around. Until there is an official notice about all completed we can't be sure.
_\|/_
U r s
ID: 884951 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 885207 - Posted: 14 Apr 2009, 2:06:50 UTC - in response to Message 884951.  

Is Astropulse 5.00 still a vaild application, or was it complete obsolated by the release of the Astropulse V_5 (5.03) ?

Astropulse v5.00 would have been needed to complete any remaining unvalidated WUs in the database, but I think they've probably all gone now - I haven't seen one for ages. ...

I would be careful with that assumption. Last week i've completed at least one of these Astropulse v5.00 wus. There might still be "a hand full" around. Until there is an official notice about all completed we can't be sure.

As they had 30 day deadlines, just 6 hosts which never report the work back would keep an AP WU alive for 90 days, add a couple which take nearly the full deadline time but don't agree...

OTOH, there's no real need anymore for an optimized AP 5.00, there will be many hosts running stock which will be prepared to do any lingering work.
                                                                  Joe
ID: 885207 · Report as offensive
Profile Steven Meyer
Avatar

Send message
Joined: 24 Mar 08
Posts: 2333
Credit: 3,428,296
RAC: 0
United States
Message 887001 - Posted: 21 Apr 2009, 21:32:40 UTC

Not a question, just an observation.

Unless someone does have an explaination handy...

I have two machines crushing S@H MB and AP with these options:

Run only the selected applications:
SETI@home Enhanced: yes
Astropulse: yes
Astropulse v5: yes

Lately is seems that the slower machine gets only AP WU, while the faster machine gets only MB cuda WU.

Where else can you say "embeecudawoo" and everyone knows what you are talking about? :^D


ID: 887001 · Report as offensive
Profile Steven Meyer
Avatar

Send message
Joined: 24 Mar 08
Posts: 2333
Credit: 3,428,296
RAC: 0
United States
Message 887006 - Posted: 21 Apr 2009, 21:35:55 UTC - in response to Message 887001.  

... two machines crushing S@H MB and AP ...


I meant to say "crunching" but then "crushing" might be appropriate too...

ID: 887006 · Report as offensive
Nessus

Send message
Joined: 27 Apr 02
Posts: 3
Credit: 1,044,992
RAC: 0
United Kingdom
Message 888021 - Posted: 24 Apr 2009, 20:44:33 UTC

Don't know if this is the correct place but I have 19 AP UNITS waiting to be processed. My average has gone from nearly 2000 to nearly 1000. I don't object to crunching AP units I would just like to get the credit due. Can anyone enlighten me as to what is going on.

ID: 888021 · Report as offensive
Profile Geek@Play
Volunteer tester
Avatar

Send message
Joined: 31 Jul 01
Posts: 2467
Credit: 86,146,931
RAC: 0
United States
Message 888028 - Posted: 24 Apr 2009, 20:56:25 UTC

They are waiting for your wingman on each WU to complete his work and return it. Then the work can be validated. AP work takes a long time to crunch and a long time in many cases for the wingman to crunch his. Many drop out of the project so the work has to time out and then be re-issued.

Think of your completed work as credits in the bank. Sometimes it takes months and months to finally get your credits but please keep on crunching and be patient. The wait is worth it.
Boinc....Boinc....Boinc....Boinc....
ID: 888028 · Report as offensive
Profile Pappa
Volunteer tester
Avatar

Send message
Joined: 9 Jan 00
Posts: 2562
Credit: 12,301,681
RAC: 0
United States
Message 888123 - Posted: 25 Apr 2009, 0:47:16 UTC - in response to Message 888021.  

Nessus

I have one machine that I only crunch Astropule on. If I look at my Pending credits it hovers around 40000 part of the reason is that it has about a Dozen Ap's waiting for people to Do Astropulse and return them (not trash them). So while the turn around for Workunits according to the Server Stats page hovers around Three Days, it does not reflect turn around time for AP results (its more like 20 days). In this case, it is patience they will get there.

Don't know if this is the correct place but I have 19 AP UNITS waiting to be processed. My average has gone from nearly 2000 to nearly 1000. I don't object to crunching AP units I would just like to get the credit due. Can anyone enlighten me as to what is going on.


Regards

Pappa

Please consider a Donation to the Seti Project.

ID: 888123 · Report as offensive
Nessus

Send message
Joined: 27 Apr 02
Posts: 3
Credit: 1,044,992
RAC: 0
United Kingdom
Message 888194 - Posted: 25 Apr 2009, 8:13:47 UTC - in response to Message 888028.  

Ok thanks for the info. I will keep crunching and just be patient. Thanks again to you and Pappa.
ID: 888194 · Report as offensive
Profile Kung

Send message
Joined: 12 Nov 01
Posts: 5
Credit: 475,350
RAC: 0
Australia
Message 889398 - Posted: 29 Apr 2009, 9:49:06 UTC

I have a question about stability. Both My AP 5.03 WU have reset back to 0% from ~75% complete, the first for the second time.

I don't get it. Is there a log file or something that will tell me what the errors are?

I am running 6.4.7 on vista SP1

My crunching does no-one any good if they keep crashing.

- Kung
ID: 889398 · Report as offensive
Previous · 1 . . . 9 · 10 · 11 · 12 · 13 · 14 · Next

Message boards : Number crunching : Astropulse FAQ


 
©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.