[RC5] farfetched stats idea?

Stephen Garrison garrison at ChE.UDel.Edu
Thu Jun 29 12:48:14 EDT 2000


One modification to the "obsolete" rating that might make it closer to
reality would be to take the average keyrate per "active" participant and
compare it to the theoretical maximum rate of a machine from the speed
page. This makes it more of an overall statistic for dnet and computers
and less of an individual stat.

Drawbacks: 

1. This means someone has to cull the speed page data and decide what
value to use for each processor. 

2. Someone needs to determine if we want to include any factor to increase
the average keyrate to take into account the machine being used by someone
else or not even being on. (I, however, think we don't make any
modification as this would give us an "lower bound on being obsolete".)

3. Some people might quit if there computer gets defined as obsolete.

Advantages:

1. It would add an interesting stat feature of "average keyrate per active
participant" allowing use to see if our growth rate is more a function of
gaining participants or gaining additional CPU power per participant from
additional or better computers.

2. It might convince people with "obsolete" computers to upgrade and,
assuming they continue with dnet, improving our overall keyrate.

-- 
Stephen Garrison

--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list