[rc5] wekly( or so) stats comparison - Bovine - Cyber
gindrup at okway.okstate.edu
Mon Oct 13 10:27:53 EDT 1997
No, actually, we don't. Before checking any keys, the reasonable
assumption was that half the keyspace would be tested before finding
Now that we have the information that the key is *not* in the first
40%, then the reasonable assumption is that half the (remaining)
keyspace will be tested before finding the key.
Extending your argument to an absurd conclusion (which I know almost
nobody would do), we would have to conclude that the expected time to
completion becomes negative after 50% is checked... But, given the
information we have about so much of the keyspace, no reasonable
person would say it is odd to not find the key in the next week. It's
somewhat disappointing for those people still carrying the expectation
from the beginning of the effort that the key would be in the first
50%, but it is not disappointing when all of the available information
collected since then is taken into account.
-- Eric Gindrup ! gindrup at Okway.okstate.edu
______________________________ Reply Separator _________________________________
Subject: Re: [rc5] wekly( or so) stats comparison - Bovine - Cyberian
Author: <rc5 at llamas.net > at SMTP
Date: 1997/10/13 09:54
On Sun, 12 Oct 1997 15:20:15 -0400 (EDT), James Mastros wrote:
>Huha? What is the difference? Since the probability for each key is
>1/2^56, then (.5*2^56)*1/2^56=.5. So when half of the keyspace is
>completed, there is a 50% chance that The Key will have been hit.
We can't count any of the keys that we have already checked when calculating
probable completion point. The current keyspace is only 60% the size of the
original one, but we still have to predict that completion will come when we
have completed 50% of the current keyspace.
"A penny saved may be a penny earned, but it's a waste of
a deposit slip and it really pisses off the tellers."
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.
More information about the rc5