score and rank


Advanced search

Message boards : Cruncher's Corner : score and rank

AuthorMessage
Tyranova
Send message
Joined: Dec 27 11
Posts: 2
Credit: 75,330
RAC: 0
Message 115735 - Posted 25 Jan 2012 10:19:07 UTC

    Last modified: 25 Jan 2012 10:23:25 UTC

    Hi,i`m the newbe here and got a question.what are the credits and ranks for? can i buy me something nice for a high score or is it just to say,look my pc is running 24/7/52? it looks a little bit like showing off,i pay an higher energybill than you..if there is nothing behind, ranks are totally sausage.i mean,the important thing is to be involved here,not to be the "best"..
    please correct me,if im wrong.

    tyranova

    Dirk Broer
    Send message
    Joined: Sep 10 05
    Posts: 9
    Credit: 6,231,323
    RAC: 1,597
    Message 115736 - Posted 25 Jan 2012 10:23:31 UTC - in response to Message 115735.

      Or is it that you are afraid you have to work through the ranks of those who are active in this project for some years and have accumulated more credit in the process than you have now?
      ____________

      Tyranova
      Send message
      Joined: Dec 27 11
      Posts: 2
      Credit: 75,330
      RAC: 0
      Message 115737 - Posted 25 Jan 2012 10:27:47 UTC - in response to Message 115736.

        no

        transient
        Send message
        Joined: Jun 3 05
        Posts: 63
        Credit: 15,824,776
        RAC: 18,677
        Message 115741 - Posted 25 Jan 2012 17:39:03 UTC

          There are two purposes, IMO. Showing of like you said, a score-card. And it could be used as a sort of temperature gauge. RAC is especially well-suited for that. It could point to a problem on your PC, overheating for example.
          ____________

          Janus
          Send message
          Joined: Nov 10 04
          Posts: 25
          Credit: 14,981,141
          RAC: 683
          Message 115742 - Posted 25 Jan 2012 18:54:17 UTC

            Some people join the project because they believe in the science goals, some people join the project because they like to be better in the scoreboard, some people join to help their team, some people join because ... ...

            I think most people here joined for a combination of many reasons. However, it doesn't really matter much why you joined. All that matters is that you did in fact join - and that we're all helping with this together in our own ways.

            That's the beauty of it all.

            Profile Gary Roberts
            Forum moderator
            Send message
            Joined: Feb 9 05
            Posts: 3018
            Credit: 1,085,646,076
            RAC: 2,356,344
            Message 115748 - Posted 26 Jan 2012 4:33:27 UTC - in response to Message 115742.

              ...All that matters is that you did in fact join - and that we're all helping with this together in our own ways.

              This is precisely the point.

              People often join out of curiosity and/or a desire to help a worthwhile 'scientific' cause. The accumulation of credit is such a seductive process that many go on to contribute far more than they would have ever thought possible when they first started.

              If you have a very small number of hosts, RAC is almost useless as a quantitative measure of performance. There will be times when your 'wingmen' are slow and you have lots of pending results and a falling RAC. There will be other times when all your results are validating quickly and RAC is rising. So the fluctuations in RAC don't really tell you about your host's performance.

              When you get 'bitten' by the bug and you build up a farm of decent proportions, the individual host fluctuations are dramatically smoothed out when averaged over the whole fleet and combined RAC can become quite stable. In a way, I'm a good example of this because I have a large number of hosts crunching Einstein pretty much exclusively. RAC becomes so stable that I can usually tell quite quickly when any significant host in the fleet goes down. Just by looking at RAC. I then look at my list of hosts and there will be one or two at the bottom that have crashed - ie not reported within the last 18 - 24 hours.

              Even though the accumulation of credit easily becomes an addiction, I would contend that there is 'good' addiction and 'bad' addiction (and probably quite a few other shades as well). I see nothing wrong with adding extra hosts and squeezing extra performance for a project that has a genuine scientific goal and a real need for those extra resources coming from volunteers. I would contend that there are a very small number of projects whose whole reason for existence is to generate huge credit, without a genuine need to solve a recognisable scientific or humanitarian problem. It saddens me when some people flock to the 'very highest payers' to the complete exclusion of the 'low paying but worthwhile' majority of projects.

              It would be quite unfair to name any project as being in the 'not worthwhile' category. That's a very subjective thing and people have to make their own judgment about that. I would contend however, that E@H is right up there in the 'worthwhile projects' list :-).

              ____________
              Cheers,
              Gary.

              Profile Bikeman (Heinz-Bernd Eggenstein)
              Forum moderator
              Project administrator
              Project developer
              Avatar
              Send message
              Joined: Aug 28 06
              Posts: 3225
              Credit: 73,541,602
              RAC: 36,834
              Message 115752 - Posted 26 Jan 2012 21:12:10 UTC - in response to Message 115748.

                Hi!

                E@H has far more to offer than credits, you can be the co-discoverer of a rare celestial body (a pulsar). I think this is pretty cool!

                HB

                ____________

                Dirk Broer
                Send message
                Joined: Sep 10 05
                Posts: 9
                Credit: 6,231,323
                RAC: 1,597
                Message 115789 - Posted 30 Jan 2012 14:19:14 UTC - in response to Message 115737.

                  no


                  Oké then! If you're into it, the beauty of statistics -of whatever nature- is that you can search for the one that best suits you (this even applies to politics).
                  I can compare myself with the members of my team -all time or on daily, weekly or monthly basis-, the people who joined up on the same date, or my fellow country men/women.
                  There is always at least one that'll show you make progress when you're active, and watching Einstein stats fills the time when you're not discovering pulsars.

                  You don't need to have two GTX 590s in SLI on a i7-3960X based socket 2011 motherboard, or a Quadro 6000/Tesla combo on a dual Xeon X5680 or dual Opteron 8262SE motherboard (though a setup like that will surely help, both for Einstein and your power company, bills will be huge).

                  I do most of my work with a humble GT 430 and an even lowlier GT 9500. It is enough not to sink away and not enough to rise rapidly, but it is good enough. My CPUs I use for other projects. I haven't discovered a pulsar yet, but with a bit of patience and another 5 years....
                  ____________

                  Post to thread

                  Message boards : Cruncher's Corner : score and rank


                  Home · Your account · Message boards

                  This material is based upon work supported by the National Science Foundation (NSF) under Grants PHY-1104902, PHY-1104617 and PHY-1105572 and by the Max Planck Gesellschaft (MPG). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the investigators and do not necessarily reflect the views of the NSF or the MPG.

                  Copyright © 2014 Bruce Allen