[rc5] Idea for team stats...

Ken Clinkenbeard ken-clinkenbeard at uiowa.edu
Sun Oct 19 14:53:47 EDT 1997

On 10/19/97, James Mastros wrote:
> We have a % of keyspace completed per team; how about a % of work completed
> done by this team (keys crunched by this team/total keys crunched)?  This
> would give the approx. % chance that this team will win, assuming that all
> teams stay at their present speed, and that the key is found by bovine.

I like the idea of having a % of keyspace stat for each team. It would be
quite humbling :-). However, it is not what you *have* crunched, but what
you *are* crunching that determines a team's chances for a "win". A team
that has been crunching for 193 days at 356 kkeys/sec has much less chance
of winning than a team that started yesterday with a keyrate of 17776
kkeys/sec even though the former team has crunched much more total
keyspace. The past performance of either team is unimportant in determing a
chance of winning as long as the winning key still lurks out there.

Your co-crunching comrade in cryptography,

 _ Ken Clinkenbeard _____________________________________________________
   Instructional Designer/Programmer
   The University of Iowa Libraries                 "Made from 100% ASCII
   TWIST Project, http://twist.lib.uiowa.edu/  for wholesome reading with
   phone:  319.335.6258                         no formatted aftertaste!"
   email:  ken-clinkenbeard at uiowa.edu
   web:  http://staffweb.lib.uiowa.edu/kclinkenbeard/

To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.

More information about the rc5 mailing list