On Fri, 02 Sep 2005 12:50:12 GMT, "Eric" <nospam@no.nospam.not> wrote:

"Louis Holleman"  wrote

Hmmmm.... this sounds familiar. Not on an oldie running an ole OS, but
on my Athlon 2 Ghz with 768 megs running Win2K. OK, that IS old....

I bet SETI Classic screamed on that though.  So, why is BOINC so
inefficient?

The same prob with the explorer. Yes peeps, it's still there.

...and extremely annoying/frusterating!

I'm gonna run another few days keeping a sharp eye onto it. If things
screw up, I'm just leaving ship. Period.

I think I'm going to mess with the CLI BOINC stuff for a bit, but if the CLI
proves to be as problematic as the GUI, then I'll be leaving the ship right
behind you.  Likewise, I've been running SETI since the start but don't feel
like continuing if it is going to require constant babysitting.   Classic
was great, efficient and never caused any problems.  I want it running
transparant and hidden in the background, requiring absolutely no hand
holding.  So far, BOINC has been everything except such.  Basically, its
just been a major PITA.

Cheers,
Eric

You must not remember the early days of Classic! It was at least as
bad as Boinc has EVER been!
Classic's problems are MANY and have been mentioned MANY times before
but basically it is a program that has outlived its usefulness and is
being replaced. There are over 50 MILLION results from Classic that
STILL need verifying to see if they actually found anything! The
program cannot track who downloads and who uploads a workunit, making
cheating a DISTINCT possibility. Cheaters HAVE been caught but it is a
laborious process hand checking the data unit by unit. Part of the
problem was the old one crunched unit means one credit. This
encouraged people that were just in it for the credits to find ways
around properly crunching and returning units. There are MANY other
things like Berkeley can't supply enough units for all of users until
they bring some more projects on-line so they keep sending out the old
units over and over and over just to keep us happy. Part of that
solution was solved when other Scientists asked if there was a way
they could jump on the bandwagon of Distributed Computing. Berkeley
was having "issues" with Classic and decided to design and write a
whole new program that would be the foundation for at least part of
their problems AND allow us users to setup our computers to crunch for
other projects that we liked. These other projects allowed Berkeley to
be able to start the process of switching from classic to Boinc. We
users would have multiple projects to choose from and would NEVER run
out of work!
Is Boinc having growing pains? ABSOLUTELY!!! Will it get better? Again
ABSOLUTELY!!! Will it be today? NO!  Will it be sooner rather than
later? HOPEFULLY!! If you can't live with these new ways of doing
things then by all means take your computers and do something else
with them, we WILL miss your help! BUT some will carry on and we WILL
make the program BETTER than Classic ever thought of being, and maybe
then you will come back. You are more than welcome to stay and help us
in our journey, but you are also welcome to stand aside and watch.