[RC5] keyrate stabilizing?
bwilson at distributed.net
Mon Oct 15 18:00:02 EDT 2001
While the number of nodes per stub is not constant, it does give us some
sense of progress. If we have 75% of the stubs back, then we can guess
that we are about 75% done with the project. Even if that last 25% of
stubs end up containing twice as many nodes (on average) than the first
75%, we at least have a gauge.
In simpler terms, the unfinished stubs represent how *many* work units
we have to do, but can't be relied on to predict how *long* it will take
us to do them.
Duplicate elimination is actually a very big problem. As I mentioned
before, some people have decided to abuse the distributed.net network by
submitting the same work over and over. Considering how long we've been
working on OGR-24, is it any surprise that we've received more work back
than it should take for the entire project, two passes and all? I don't
know this for sure, but I do know all the work for OGR-24 was handed out
more than twice.
We've tried using raw "unfiltered" statistics before (see CSC), and the
response was not pretty. Talk about conspiracy theories! I have a hard
time believing a stats page showing "OGR-24 is 274% complete" (or
whatever it is) would be an improvement over what we have now.
Bruce Wilson <bwilson at distributed.net>
PGP KeyID: 5430B995, http://www.toomuchblue.com/
"Quidquid latine dictum sit, altum viditur."
(Whatever is said in Latin sounds profound.)
| -----Original Message-----
| total stubs returned / total stubs * 100 = rough % done.
| It seems it would be much easier to give a rough percentage complete
| figure. Checking for duplicates and checking the 2nd pass to
| ensure acuracy
| of the returned data does indeed need to be done at some
| stage, but for a
| simple 'we are about 74% done' figure, couldn't we just count
| the number of
| returned stubs? Surely duplicate results are neglible in
| comparision to
| valid results and thus would not distort the rough figure too much.
| The other inaccuracy in this figure is that each stub
| contains a different
| number of nodes. So a stub is not an accurate measure of
| work done. But
| doesn't this average out in the long term, or does the
| average number of
| nodes per stub deviate as we move though the workspace?
| To unsubscribe, send 'unsubscribe rc5' to
| majordomo at lists.distributed.net
| rc5-digest subscribers replace rc5 with rc5-digest
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest
More information about the rc5