Mine is 11 months long. I suppose this is to allow for slower computers. So it could be months before I get credit? I just hope we return valid results.


edit: I just did some calculations, and my laptop here at work that I am using is only on for 8 hours or so a day. According to BOINC, I have 784 hours to go which would mean that it should get done in 100 days or sooner. I guess I won't be bugging the Seti or Predictor servers for a while, :)

Lol, you'll get credit when you finished the run ;)

> Mine is 11 months long. I suppose this is to allow for slower computers. So it
> could be months before I get credit? I just hope we return valid results.
>
For me this is a real negative point to having CP on my machine. Other projects in which I choose to participate in get hosed by CP's huge units of work.

I don't pretend to understand this project as of yet, but I'm curious to see if smaller WUs could be generated?

you actually get "incremental credit" every trickle, so it won't seem as bad!
:-)

> For me this is a real negative point to having CP on my machine. Other
> projects in which I choose to participate in get hosed by CP's huge units of
> work.
>
> I don't pretend to understand this project as of yet, but I'm curious to see
> if smaller WUs could be generated?

Remember that SETI's small WUs are created by splitting up vast quantities of data into little chunks. Credit is granted according to the processing work done. A month's worth of SETI = the same amount in CPDN. And with CPDN you do get credit as you go (several times a day for faster machines) - you don't have to wait for others to duplicate your work, unlike other projects. The current version of the client allows you to share between projects, so one project isn't pushing out another unless you want it to, though you will have to be patient and give it time to settle down.

It isn't practicable to split CPDN WUs into smaller ones. You are running a <b>big</b> climate model (around a million lines of code) and trying to model the earth's climate over 45 years. It generates vast amounts of data in the process, and splitting the model up would mean shifting that data around, creating all sorts of bandwidth problems for participants as well as the project. Remember that BOINC, like other DC efforts, is designed to run in the background for months or years, so it doesn't make any difference in that respect whether it is running a few big WUs or lots of smaller ones (except that CPDN requires less data transmission).

The good news for those of us who are excited by the challenge of running programs originally designed for supercomputers is that there will be an opportunity to participate in even bigger tasks from CPDN in the coming months. Once you've been running CPDN for a year you start wanting something you can <b>really</b> get you teeth into!