[RC5] OGR motivations/achievements in general
bwilson at distributed.net
Wed Oct 9 18:33:38 EDT 2002
No, what you're running now is not a waste of time. As I mentioned in
my third paragraph, we are doing what we can to avoid handing out work
that has already been completed once. While roughly 60% of the work we
hand out is finished each time through, that means about 40% is not
There is no client available today which is not vulnerable to the
truncation bug. However, every client since about 463 (?) can do work
which is believed to be useful for Pass 1. When a "fixed" OGR client is
released, it will be used to confirm or discard Pass 1 work. Until we
get that client out, we are stuck on Pass 1, but Pass 1 is not done yet
Bruce Wilson <bwilson at distributed.net>
PGP KeyID: 5430B995, http://www.toomuchblue.com/
Build a man a fire and he'll be warm for a day.
Set a man on fire, he'll be warm for the rest of his life.
| -----Original Message-----
| From: owner-rc5 at lists.distributed.net
| [mailto:owner-rc5 at lists.distributed.net] On Behalf Of blitz
| Sent: Wednesday, October 09, 2002 15:43
| To: rc5 at lists.distributed.net
| Subject: RE: [RC5] OGR motivations/achievements in general
| Woah..you mean what we're running client wise is a waste of time?
| The latest clients? Is there a "proper" one for Linux and
| Windoze available
| that IS ok to use?
| I'd rather not have to revisit all the machines in my lan to
| change until
| we've got a keeper for a client.
| Looks like Ive got to visit them now and shut them down, I
| don't think I
| want to have them running defective output that will be
| discarded anyway.
| Is .473 ok?
| At 13:45 10/9/02 -0500, you wrote:
| >Because of the nature of the bug, if the ruler is truncated once, it
| >will be truncated every time. Unfortunately, this means we will be
| >discarding essentially all 2nd pass work with existing
| clients. (sigh -
| >I hate giving bad news).
| >Again, this is to the best of my knowledge - I am not a
| coder for OGR,
| >and I don't know all the details, I am merely echoing what I
| >from discussions on IRC.
| >We have from time to time gone over the work received and updated the
| >keymaster's maps for which workunits have not yet been
| received, in an
| >effort to make the most of the work being handed out. (No
| sense handing
| >it out again if we already have one apparently valid
| result). Like our
| >other projects, the new-work-return rate has declined and the
| >duplicate-work rate has increased the closer we get to the end, but
| >there is still a significant percentage of work not yet received back
| >for the first time.
| >Bruce Wilson <bwilson at distributed.net>
| >PGP KeyID: 5430B995, http://www.toomuchblue.com/
| >Build a man a fire and he'll be warm for a day.
| >Set a man on fire, he'll be warm for the rest of his life.
| >| -----Original Message-----
| >| From: owner-rc5 at lists.distributed.net
| >| [mailto:owner-rc5 at lists.distributed.net] On Behalf Of Jeff Leigh
| >| Sent: Wednesday, October 09, 2002 01:34
| >| To: rc5 at lists.distributed.net
| >| Subject: Re: [RC5] OGR motivations/achievements in general
| >| I understand the need to verify the current results with
| the new (as
| >| yet unavailable) client, but it seems like you should already have
| >| gotten thru all the stubs several times over by now.
| >| Here is Dave Avery's .plan file from 2000:
| >| :: 23-Sep-2000 04:14 (Saturday) ::
| >| OGR progress report:
| >| As of 23:59 22-sept-2000 there were 2,264,082 unique OGR-25 stubs
| >| returned,
| >| out of 12,207,683, or 14.89% complete
| >| As of 23:59 22-sept-2000 all but 7 unique OGR-24 stubs had
| >| been returned
| >| and about 49% had valid pass-2 returns
| >| So over 2 years ago, OGR-24 was already half way thru the
| second pass
| >| and OGR-25 was already at 15% up from 9% on Sept 11th. So how many
| >| times have these stubs been checked over with the 'broken' client?
| >| Seems like every stub would have been checked numerous
| times by now.
| >| On Friday, October 4, 2002, at 09:16 AM, Bruce Wilson wrote:
| >| > Because of the verification plan we are using, we do not
| >| expect to have
| >| > to discard work done with current buggy clients unless the
| >| bug affected
| >| > the results for a given stub. Our original plan was to
| require two
| >| > identical results for the same stub from different
| >| participants. Now
| >| > we
| >| > will be revising that to say that one of the results
| must also come
| >| > from
| >| > a "fixed" client. If the buggy client gives the same
| results as the
| >| > fixed client, the bug was not a factor. If the buggy
| client gives a
| >| > different result, then we wait for a confirmation from a
| >| > participant using the fixed client.
| >| --
| >| 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
| 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