seti results

[david]

I guess this is the section to post this(sorry you get no fancy graphics with me just the meat and potatoes)

        Whats the procedure for the SETI results to go from claimed to granted.?

         Normally how long does that process take and what are the criteria for granting?

          Reason for question i have quite a number of PENDING credits going back several days,i certainly can't see any reason why they wouldn't be granted.

          I must confess i don't really understand how the points system works.

          Would appreiate any enlightenment

[UBT - Timbo]

I guess this is the section to post this(sorry you get no fancy graphics with me just the meat and potatoes)

No worries

Whats the procedure for the SETI results to go from claimed to granted.?

The results you return form part of a quorum - that is: the same original WU is sent to multiple crunchers who then return the work. Once the specified number of results are returned (and for SETI that number is 3) and (roughly) compared to each other, the credit is issued, based on the 3 different returned results.

Normally how long does that process take and what are the criteria for granting?

The completion deadline for a SETI WU can vary - for some WUs it's a month. So, if you return a result very quickly, and two other members take longer to complete the WU, then you could be waiting up to a month. If no-one returns a result, then you could be waiting until the same WU is sent out again and then returned. In a worse case scenario, you could be waiting for ever...(as no-one (except you) ever returns a completed result for this WU!)

Reason for question i have quite a number of PENDING credits going back several days,i certainly can't see any reason why they wouldn't be granted.

I wouldn't worry about it....once the credits start flowing, as long as you keep crunching, then your credits total will start rising quite quickly. Just have to be patient to start with !!

I must confess i don't really understand how the points system works.

Nor does anyone else !! It's mainly to do with your CPU and it's benchmarking (when you start up the BOINC client it tests your CPU to see how much grunt it has !)

Once at least 3 results are returned (and each claims a specific amount of credit) then the credit awarded is mathematically linked to the 3 different claims for credit (It used to be the average of the 3).    

Would appreiate any enlightenment

Hopefully, this will give you some !

regards

Tim

[david]

Ahaa said the blind man whilst reading the paper to his deaf son,"i see" tis as clear as mud.

            Thanks Tim that makes a lot more sense i didn't realise that each WU was sent out 3 times,logically that makes sense, my only concern was that you may have lost your credits after say a week or two and the granting of credit seemed somewhat random.

             I think the light at the end of that tunnel has been turned back on.

[UBT - Timbo]

Ahaa said the blind man whilst reading the paper to his deaf son,"i see" tis as clear as mud.

            Thanks Tim that makes a lot more sense i didn't realise that each WU was sent out 3 times,logically that makes sense, my only concern was that you may have lost your credits after say a week or two and the granting of credit seemed somewhat random.

             I think the light at the end of that tunnel has been turned back on.

Oh, I do so like turning something on !.....

If you go to your own "My Account" link on SETI@home website and then click on "Results" and then click on one of the numbers under the "Work Unit ID" heading, you can see the different computers that have crunched the same WU as you. Some may have completed the work (and you cana see the credits they have claimed, as well as your own claim), others may still be in progress.

It's unlikely you'll lose any credits as long as they are validated and returned before the deadline.

regards,

Tim

[david]

Yeah those light switchs can be damnably tricky at times,still its all in the way you flickem.

     I had gone in and had a look at that before however didn't realise that yhere were 4 units the same with a quorum of 3,if i had paid attention and thought about it i would have understood it,the attention span of a ferret.  

      All quite logical and predictable when you think on it,by jove berkeley must have a sizable mainframe to deal with all the crunching they receive fromthe crunchers,potentially an awful lot of toothaches if not managed correctly.

[UBT - Timbo]

All quite logical and predictable when you think on it,by jove berkeley must have a sizable mainframe to deal with all the crunching they receive fromthe crunchers,potentially an awful lot of toothaches if not managed correctly.

In the early days of BOINC, SETI@home had all manner of issues.....most of it was hardware related....us grey-haired old folk remember those days fondly....! Sometimes you really didn't know whether the project would survive - and they still have cash-flow problems even now, despite many donating cash to the project.

By current standards, most of their hardware is quite antiquated....but as most of it runs *nix or Solaris etc, so it can still do a lot more than you'd think. They have just stuffed it with memory and large hard drives and it works to a fashion. (If you search around their site, you might even come across some of their server room photo's...start here: http://setiathome.berkeley.edu/sah_photos.php !!)

This is off their site tonight:

Hosts:
jocelyn: Sun V40z (2 x 1.8GHz Opteron, 16 GB RAM)
sidious: Intel Server (3 x dual core 3GHz Xeon, 16 GB RAM)
thumper: Sun Fire X4500 (2 dual-core 2.6GHz Opteron, 8 GB RAM)
bruno: Intel Server (2 x 2.8GHz Xeon, 12 GB RAM)
kosh: Sun D220R (2 x 440MHz Sparc, 2 GB RAM)
klaatu: Sun D220R (2 x 440MHz Sparc, 2 GB RAM)
koloth: Sun D220R (2 x 440MHz Sparc, 1 GB RAM)
kryten: Sun E3500 (6 x 400MHz Sparc, 6 GB RAM)
galileo: Sun E3500 (8 x 400MHz Sparc, 7 GB RAM)
penguin: Sun D220R (2 x 440MHz Sparc, 2 GB RAM)

(Those Sun units are really ancient !!) Course, if they could afford a supercomputer, then they probably wouldn't need us !!

regards

Tim

[david]

Thanks Tim that is quite fascinating, i had pictured it as a typical American set up full of bells and whistles and b/s,however it really is pretty archaic,but it obvivously works,i will peruse at my leisure,i may even learn a fact or three,not terribly difficult.

          I had the old SETI,before BOINC but when boinc came along i could never get it to work properly on my m/c,shelved it ,forgot about it and eventually deleted it.

          I then wanted the challenge of getting it to work on Linux which,i eventually did,with the invaluable help of John.When i restarted it i certainly didn't realise that there were as many projects as are avalible,so i will have a look at that shortly.

          Please excuse my typing skills,on reading back through them they are quite abominable,they are on par with an arthritic neanderthal.

[UBT - Timbo]

...I had the old SETI,before BOINC but when boinc came along i could never get it to work properly on my m/c,shelved it ,forgot about it and eventually deleted it.......

So, do you recall your original BOINC username and even the original email address you used to register with...(assuming they are different to your "new" username).

If so, then your stats are still around on one of the stats websites....!

Might be worth having a look for them... .

And depending on which project you were crunching at the time, you may still be able to link to them....

regards,

Tim

[david]

Hello Tim, yes i know the e-mail address that i used then,can't recall the password.

         Where would you begin to search for them,if they still exist its probably 3 years ago or more now.

          I guess the next thought is are those results transferrable to credits and be added to your current credits

         (Must confess it would feel a bit like cheating,but if it is all above board,well..............

[UBT - Timbo]

Hello Tim, yes i know the e-mail address that i used then,can't recall the password.

As long as you still have access to your old email account (i.e. you can still retrieve emails sent to that old address) that you used, then you can get the project to send you the password....

Where would you begin to search for them,if they still exist its probably 3 years ago or more now.

You need to be able to remember your old username....then just use this link:

http://www.boincsynergy.com/stats/index ... earchteam=

and in the top right box under "Search for Member Name:" just put in your old username (where "ubt" is) - then you'll see which accounts are held with each project. Of course some people may also have chosen the same name, but at least it migth remind you of which project you crunched on before.

(And of course, not every project available today was around 3 years ago....I think there was only SETI, CPDN, Einstein, Rosetta and maybe Predictor !)

I guess the next thought is are those results transferrable to credits and be added to your current credits

(Must confess it would feel a bit like cheating,but if it is all above board,well..............

All you need do, is to log in to your old account(s), change the username and email address to your current username/email and then they will "line up". The only one(s) that won't is any project you have now joined, which you joined 3 years ago, as member accounts under two different email addresses cannot be merged. So, say your old SETI account had 1,000's of credits, I'd dump your new account and re-start crunching on your old account....

regards,

Tim

[david]

RAIN GLORIOUS RAIN  woke up this morning to the sound of beautiful soft rain falling so went for a bit of a walk in it ,wonderful,maybe the drought has broken,sure hope so,must be the naked rain dance i did when i finally got SETI working or perhaps the ritual fertility rite with next doors cat.

            Who cares it worked incidentally you may find it a little odd waxing lyrical over something as mundane as rain but we have little rain for the last 5/6 years,informed sources suggest this is the worst drought for a 1000 years.  

             So send it down HUGHY(thats an old Australian saying to do with rain)

             Sorry now back to the matter in hand.Tim i had a look around inside that url you posted it didn't recognize anything i put in ,i'll have a search around for my old p/w and continue to attempt to locate the old figures i usen't to pay a great deal of attention to them SETI just ran in the background, i wasn't part of a team or even took that much notice of it.

              Can't say i understand exactly what you suggest is the best way to go but will cross that bridge if and when we get to it.

[UBT - Timbo]

....Sorry now back to the matter in hand.Tim i had a look around inside that url you posted it didn't recognize anything i put in ,i'll have a search around for my old p/w and continue to attempt to locate the old figures i usen't to pay a great deal of attention to them SETI just ran in the background, i wasn't part of a team or even took that much notice of it.

Can't say i understand exactly what you suggest is the best way to go but will cross that bridge if and when we get to it.

No worries.

If I can be of any further help, just post a message in the forum...!

regards,

Tim