climateprediction.net (CPDN) home page
Thread 'My stats page at BOINCstats/BAM is totally useless.'

Thread 'My stats page at BOINCstats/BAM is totally useless.'

Message boards : Number crunching : My stats page at BOINCstats/BAM is totally useless.
Message board moderation

To post messages, you must log in.

AuthorMessage
EveningStarNM

Send message
Joined: 27 Oct 13
Posts: 14
Credit: 29,828
RAC: 0
Message 50129 - Posted: 11 Sep 2014, 17:04:09 UTC
Last modified: 11 Sep 2014, 17:07:08 UTC

It will take months to smooth out that erroneous -- and gigantic -- bump. I am not a happy camper. I enjoyed the competition I was having with some friends, but now, instead of having BOINCstats/BAM do the comparisons for us, especially with regard to recent average credit, we'd have to do the calculations by hand for my account. Unfortunately, that's too much work for a casual thing like this, and I have to drop out of the competition.

It's not about the credit. It's about the fun in competing with friends -- even though I was never even close winning until CPDN totally f*****d up my stats. What's worse is that this seems to have been happening to others for a while.

Bummer.

Please fix it.
ID: 50129 · Report as offensive     Reply Quote
Eirik Redd

Send message
Joined: 31 Aug 04
Posts: 391
Credit: 219,896,461
RAC: 649
Message 50130 - Posted: 11 Sep 2014, 18:09:26 UTC - in response to Message 50129.  
Last modified: 11 Sep 2014, 18:10:07 UTC

It will take months to smooth out that erroneous -- and gigantic -- bump. I am not a happy camper. I enjoyed the competition I was having with some friends, but now, instead of having BOINCstats/BAM do the comparisons for us, especially with regard to recent average credit, we'd have to do the calculations by hand for my account. Unfortunately, that's too much work for a casual thing like this, and I have to drop out of the competition.

It's not about the credit. It's about the fun in competing with friends -- even though I was never even close winning until CPDN totally f*****d up my stats. What's worse is that this seems to have been happening to others for a while.

Bummer.

Please fix it.


I, too, was there -- when the people looking at the air and water totally missed the silicon problem :)

We humans are such limited losers.

The stats sites will eventually average out. "Believe me" TANSTAFFL I usta bitch about this kinda stuff.

Now, I only hope that Oxford gets funded, and my grandkids don't need to depend on their biochem PHd's to even survive the next century.
ID: 50130 · Report as offensive     Reply Quote
EveningStarNM

Send message
Joined: 27 Oct 13
Posts: 14
Credit: 29,828
RAC: 0
Message 50131 - Posted: 11 Sep 2014, 18:36:30 UTC - in response to Message 50130.  

Unfortunately, our descendants /will/ have to rely on their PhDs in a whole slew of sciences, some of which haven't even been invented yet, in order to survive -- assuming, of course, that we stop dumbing down our education system and give them one they can use. But until we get our political system cleaned up and wrest control of it from the corrupt elite, there's not much we can do for them except nibble away at the edges with tiny efforts like this. Sadly, that doesn't seem likely to happen any time soon. So I'm just trying to have a little fun wherever I can find it. This friendly bet, which I was destined to lose, was a small pleasure, but an important one. (Ironically, I'm the sysadmin in the group and have more computing power available than any of them -- which made me the butt of endless jokes).
ID: 50131 · Report as offensive     Reply Quote
SandJ

Send message
Joined: 4 Jan 07
Posts: 9
Credit: 575,832
RAC: 0
Message 50134 - Posted: 11 Sep 2014, 21:29:17 UTC - in response to Message 50129.  

It will take months to smooth out that erroneous -- and gigantic -- bump. I am not a happy camper. I enjoyed the competition I was having with some friends, but now, instead of having BOINCstats/BAM do the comparisons for us, especially with regard to recent average credit, we'd have to do the calculations by hand for my account. Unfortunately, that's too much work for a casual thing like this, and I have to drop out of the competition.

So make it fun. Declare yourself "Ultra �ber-winner Of All Time 2014", change your .sig or whatever accordingly permanently, and generously announce you have bowed out of the competition so the lesser mortals stand a chance. ;-)

I know it is annoying, but of all the shit that can happen on an 11th of September, this is possibly not the worst. After all, you might not have been given any credits at all for work done; now that does make my blood boil.
ID: 50134 · Report as offensive     Reply Quote
EveningStarNM

Send message
Joined: 27 Oct 13
Posts: 14
Credit: 29,828
RAC: 0
Message 50136 - Posted: 12 Sep 2014, 1:17:01 UTC - in response to Message 50134.  
Last modified: 12 Sep 2014, 1:21:32 UTC

After all, you might not have been given any credits at all for work done; now that does make my blood boil.


That actually happened to me with a couple of tasks on another project a while back. It was wonderful. I got a lot of mileage out of it with my friends. I was far behind them (I'm using three old machines for BOINC while they're using their latest and greatest single machines -- and kicking my ass), and it gave me something to fend off the teasing. They didn't make me buy the drinks that week even though they knew I was lying my ass off about how much credit I should have gotten.

Now that's fun! But this is going to cost me big time.
ID: 50136 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 50140 - Posted: 12 Sep 2014, 3:49:45 UTC - in response to Message 50136.  

Sorry, but this project isn't about playing games with credits. It's wholly about climate modelling.

The current problem has been fixed, although why the script went so badly wrong still isn't known yet.

Nothing further can be done about credits on external sites, but you could start running models again.
There's still plenty of the new hadcm3 "short" models, which take less than 24 hours to run, but have two huge zips.

ID: 50140 · Report as offensive     Reply Quote
EveningStarNM

Send message
Joined: 27 Oct 13
Posts: 14
Credit: 29,828
RAC: 0
Message 50142 - Posted: 12 Sep 2014, 6:23:24 UTC - in response to Message 50140.  
Last modified: 12 Sep 2014, 6:32:49 UTC

Sorry, but this project isn't about playing games with credits. It's wholly about climate modelling.

The current problem has been fixed, although why the script went so badly wrong still isn't known yet.

Nothing further can be done about credits on external sites, but you could start running models again.
There's still plenty of the new hadcm3 "short" models, which take less than 24 hours to run, but have two huge zips.


I'll make a deal with you: You live your life your way, and I'll live my life my way. Does that sound okay with you? But maybe I'll risk wasting CPU power on this project again. I had a lot of trouble with failed downloads and computing errors last year, so I gave up on CPDN and started giving computer time to projects that didn't exhibit such problems. I'll load a few tasks and see what happens.
ID: 50142 · Report as offensive     Reply Quote
EveningStarNM

Send message
Joined: 27 Oct 13
Posts: 14
Credit: 29,828
RAC: 0
Message 50143 - Posted: 12 Sep 2014, 7:39:02 UTC - in response to Message 50140.  
Last modified: 12 Sep 2014, 7:41:53 UTC

Sorry, but this project isn't about playing games with credits. It's wholly about climate modelling.


I should have conceded that, of course, you're right -- in exactly the same way that singing in the cotton fields wasn't the reason the slaves were picking cotton, but they sang anyway.
ID: 50143 · Report as offensive     Reply Quote
EveningStarNM

Send message
Joined: 27 Oct 13
Posts: 14
Credit: 29,828
RAC: 0
Message 50146 - Posted: 12 Sep 2014, 15:29:41 UTC - in response to Message 50140.  
Last modified: 12 Sep 2014, 15:30:44 UTC

...you could start running models again.
There's still plenty of the new hadcm3 "short" models, which take less than 24 hours to run, but have two huge zips.


Well, I tried again. I downloaded four tasks to run on a fresh and clean installation of Windows Server 2008 R2 and BOINC 7.2.42. I did upgrade VirtualBox from the bundled version to 4.3.12 because the old version kept crashing all three of the machines I ran it on, resulting in BSODs. But, so far, two of the tasks have resulted in "errors while computing". See 17020637 and 17020630. Both tasks died with too many "INITTIME: Atmosphere basis time mismatch" errors.

I don't believe in waste of any kind, including CPU cycles, so I'll go back to the projects I trust.
ID: 50146 · Report as offensive     Reply Quote
ProfileIain Inglis
Volunteer moderator

Send message
Joined: 16 Jan 10
Posts: 1084
Credit: 7,808,726
RAC: 5,192
Message 50148 - Posted: 12 Sep 2014, 16:57:40 UTC - in response to Message 50146.  

...I don't believe in waste of any kind, including CPU cycles, so I'll go back to the projects I trust.

You could limit yourself to model types you trust: EU, ANZ, PNW - all fine.
ID: 50148 · Report as offensive     Reply Quote
EveningStarNM

Send message
Joined: 27 Oct 13
Posts: 14
Credit: 29,828
RAC: 0
Message 50149 - Posted: 12 Sep 2014, 17:45:32 UTC - in response to Message 50148.  
Last modified: 12 Sep 2014, 17:50:38 UTC

You could limit yourself to model types you trust: EU, ANZ, PNW - all fine.

If I'm not mistaken, I'd have to manually cancel the tasks that I don't trust and hope that I'll download one of those that you mentioned, right? I don't see -- and I don't expect -- anything in the project preferences that allow such fine-grained control of model selection.

The thing I like about BOINC is that, if I pick the right projects, I don't have to manage it. I don't want to have to manage it. I just want it to work without crashing my machines or wasting their time.

I'm still running a 180-hour ANZ task and a 30-hour HADCM3S 1HAB task. Frankly, I'll be surprised, but delighted, if they don't end in errors. We'll see.

Meanwhile, all 1CFP tasks should be removed from the project and not distributed. They don't appear to return good results ever.
ID: 50149 · Report as offensive     Reply Quote
Professor Desty Nova
Avatar

Send message
Joined: 19 Sep 04
Posts: 92
Credit: 2,011,637
RAC: 351
Message 50151 - Posted: 12 Sep 2014, 20:42:41 UTC - in response to Message 50149.  
Last modified: 12 Sep 2014, 21:18:35 UTC

You could limit yourself to model types you trust: EU, ANZ, PNW - all fine.

If I'm not mistaken, I'd have to manually cancel the tasks that I don't trust and hope that I'll download one of those that you mentioned, right? I don't see -- and I don't expect -- anything in the project preferences that allow such fine-grained control of model selection.


Go to your account here on climateprediction.net, and click "climateprediction.net preferences", edit them, and on "Run only the selected applications", select the applications you want to run.

On your BOINC Manager Client, click update.

PS: This is not unique to climateprediction. Lots of other BOINC projects let you choose their respective applications when there are more than one.


Professor Desty Nova
Researching Karma the Hard Way
ID: 50151 · Report as offensive     Reply Quote
EveningStarNM

Send message
Joined: 27 Oct 13
Posts: 14
Credit: 29,828
RAC: 0
Message 50152 - Posted: 12 Sep 2014, 22:16:30 UTC - in response to Message 50151.  

Go to your account here on climateprediction.net, and click "climateprediction.net preferences", edit them, and on "Run only the selected applications", select the applications you want to run.

On your BOINC Manager Client, click update.

PS: This is not unique to climateprediction. Lots of other BOINC projects let you choose their respective applications when there are more than one.


Thanks! I don't play much with BOINC settings, and I'd never noticed that. I've just been letting it do its thing, intervening only when something crashed.
ID: 50152 · Report as offensive     Reply Quote

Message boards : Number crunching : My stats page at BOINCstats/BAM is totally useless.

©2024 cpdn.org