How many credits is worth one workunit ???


log in

Advanced search

Message boards : Number crunching : How many credits is worth one workunit ???

Author Message
kikiman
Send message
Joined: 21 Jan 03
Posts: 2
Credit: 2,272
RAC: 0
Poland
Message 28116 - Posted: 19 Sep 2004, 19:46:33 UTC

because I'm still wondering and I can't figure it out.
Sorry for my bad English :)

Profile Paul D. Buck
Volunteer tester
Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 28155 - Posted: 19 Sep 2004, 21:33:57 UTC - in response to Message 28116.

> because I'm still wondering and I can't figure it out.
> Sorry for my bad English :)

Well, I cannot speak any other language besides FORTRAN at all, so don't feel bad, besides, I like your accent!

:)

Ok, the credit granted is based on processing time vs. the speed of an "ideal" defined reference computer. Bottom line, it depends ...

To make it more complicated we have policy rules that say if you get the max requested credit, min, or an average (or the median value) ...

In my BOINC FAQ there is a whole long bit aboiut credit... Also, in the glossary you can look up credit, related terms and concepts.
<p>
For BOINC Documentaion: Click Me!


Profile Jaaku
Volunteer tester
Avatar
Send message
Joined: 29 Oct 02
Posts: 494
Credit: 346,224
RAC: 0
United Kingdom
Message 28157 - Posted: 19 Sep 2004, 21:44:11 UTC

In Uncomplicated words, the Wu you are ding has to be done by 2 other people, the middle value of all 3 will be the granted credit for all of you.

EG: person1 = 4, person2 = 6 , person3 = 15 you will all be granted 6.

Now you asked how does 'it' work it our, Boinc just dishes out credit for how good the WU was and the time taken!

Jaaku

^ Click Here ^ ^ Click Here ^ ^ Click Here ^

jjhat1
Send message
Joined: 24 Apr 03
Posts: 49
Credit: 61,357
RAC: 0
United States
Message 28210 - Posted: 20 Sep 2004, 0:41:10 UTC - in response to Message 28116.

In a perfect world each work unit should request the same amount of credit no matter how processes it.

It is the work unit that determines the amount of credit and not who processes it.

In the perfect world every work unit would be processed by each computer give the same result and ask for the same amount of credit. However, because different systems use different amounts of numbers after the decimal and some round and some truncate we get different results. Also having different speed computers with only rough benchmarks for comparison makes it differ some as well.

kikiman
Send message
Joined: 21 Jan 03
Posts: 2
Credit: 2,272
RAC: 0
Poland
Message 28281 - Posted: 20 Sep 2004, 8:47:37 UTC

wow, I thought it would be less complicated but I was wrong
Thanks for all of you for help :))))

Profile Toby
Volunteer tester
Avatar
Send message
Joined: 26 Oct 00
Posts: 1005
Credit: 5,604,056
RAC: 2,014
United States
Message 28285 - Posted: 20 Sep 2004, 9:03:41 UTC

Also, there are different types of work units. Some take longer or shorter depending on how fast/slow the telescope was moving while the work unit was being recorded. And then of course there are work units with too much noise that cancel out before they are 'done'. So yes, complex it is :)


---------------------------------------
- A member of The Knights Who Say NI!
Possibly the best stats site in the universe:
http://boinc-kwsn.no-ip.info

Profile Paul D. Buck
Volunteer tester
Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 28295 - Posted: 20 Sep 2004, 9:32:47 UTC - in response to Message 28285.

> Also, there are different types of work units. Some take longer or shorter
> depending on how fast/slow the telescope was moving while the work unit was
> being recorded. And then of course there are work units with too much noise
> that cancel out before they are 'done'. So yes, complex it is :)

There is a handy chart and some results vs. other things (provided by Comotose) that talks to this and other issues. That is in the SETI@Home FAQ I wrote.

Credit and all the wonders about it are discussed in the BOINC FAQ, touched on in the SETI@Home FAQ and in the glossary where the terms are defined.
<p>
For BOINC Documentaion: Click Me!


Profile Heaphus
Volunteer tester
Send message
Joined: 1 Apr 03
Posts: 96
Credit: 3,070,830
RAC: 0
United States
Message 28398 - Posted: 20 Sep 2004, 16:04:44 UTC - in response to Message 28295.


>
> There is a handy chart and some results vs. other things (provided by
> Comotose) that talks to this and other issues. That is in the SETI@Home FAQ I
> wrote.
>

Could you provide a more direct link to this information? Thanks.

Profile Paul D. Buck
Volunteer tester
Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 28608 - Posted: 21 Sep 2004, 11:39:06 UTC - in response to Message 28398.

>
> >
> > There is a handy chart and some results vs. other things (provided by
> > Comotose) that talks to this and other issues. That is in the SETI@Home
> FAQ I
> > wrote.
> >
>
> Could you provide a more direct link to this information? Thanks.

As part of an "unofficial" effort to document the SETI@Home Powered by BOINC, the following documents have been written to aid you in the use of SETI@Home Powered by BOINC:

Documentation Menu (pdb)
Release Notes =

Message boards : Number crunching : How many credits is worth one workunit ???

Copyright © 2014 University of California